[Jason_W._Neyers,_Richard_Bronaugh,_Stephen_G._A._(BookFi.org).pdf

JOBNAME: Neyers - Exploring C PAGE: 2 SESS: 5 OUTPUT: Thu Apr 16 10:01:56 2009

Columns Design Ltd / Job: Neyers / Division: Prelims /Pg. Position: 2 / Date: 26/2


Exploring Contract Law

Edited by

Jason W NeyersRichard Bronaugh and

Stephen G A Pitel



Published in North America (US and Canada) byHart Publishing

c/o International Specialized Book Services920 NE 58th Avenue, Suite 300

Portland, OR 97213–3786USA

Tel: +1 503 287 3093 or toll-free: (1) 800 944 6190Fax: +1 503 280 8832

E-mail: [email protected]: http://www.isbs.com

© The editors and contributors severally 2009

The editors and contributors have asserted their right under the Copyright, Designs andPatents Act 1988, to be identified as the authors of this work.

All rights reserved. No part of this publication may be reproduced, stored in a retrievalsystem, or transmitted, in any form or by any means, without the prior permission of HartPublishing, or as expressly permitted by law or under the terms agreed with the appropriatereprographic rights organisation. Enquiries concerning reproduction which may not becovered by the above should be addressed to Hart Publishing at the address below.

Hart Publishing Ltd, 16C Worcester Place, Oxford, OX1 2JWTelephone: +44 (0)1865 517530 Fax: +44 (0)1865 510710

E-mail: [email protected]: http://www.hartpub.co.uk

British Library Cataloguing in Publication DataData Available

ISBN: 978-1-84113-906-7

Typeset by Columns Design Ltd, ReadingPrinted and bound in Great Britain by

TJ International Ltd, Padstow, Cornwall



EXPLORING CONTRACT LAW

In this book, leading scholars from Australia, Canada, Hong Kong, NewZealand, Singapore, the United Kingdom and the United States deal withimportant theoretical and practical issues in the law of contract andclosely-related areas of private law. The chapters analyse developments inthe law of estoppel, mistake, undue influence, the interpretation ofcontracts, assignment, exclusion clauses and damages. The chapters alsoaddress more theoretical issues such as discerning the limits of contractlaw, the role of principle in the development of contract doctrine and themorality of promising. With its rich scope of contributors and topics,Exploring Contract Law will be highly useful to lawyers, judges andacademics across the common law world.



Foreword

The chapters in this volume are the product of a symposium titled‘Exploring Contract Law’ held at the University of Western OntarioFaculty of Law in January 2008. The sessions featured good intellectualpunch-ups between those participants who see law as an adjectival studyfocused on the work of the courts and commerce, and others who dedicatetheir careers to more philosophical musings about legal concepts. Bothapproaches are helpful. Problem solving without a sound philosophicalbasis risks palm tree justice. Theory without a nod to practice risksirrelevance. Those in the first group would likely agree with KarlLlewellyn’s definition of law as what legal people do:

This doing of something about disputes, this doing of it reasonably, is thebusiness of law. And the people who have the doing in charge, whether they bejudges or sheriffs or clerks or jailers or lawyers, are officials of the law. Whatthese officials do about disputes is, to my mind the law itself.1

However, chapters in this volume by those in the second group show themto be in determined pursuit of what might be called a pure theory ofcontract law. At the symposium, one of the participants conjured up (insupport of the argument that contracts are not necessarily promises) anelectrical repairman who includes in his wiring contract a rather unlikelyclause:

The commitments expressed herein are exclusively contractual. We intend herebyto bind ourselves contractually to make the payments and to perform the actsspecified, but we do not intend to bind ourselves morally to do so: these arecontractual undertakings, not promises (emphasis added).

I do not expect to see such contractual provisions anytime soon and thetheoreticians probably have no expectation that we will, or even muchinterest in whether we do. Such writings constitute ‘an exercise in logic, notin life,’2 and derive their nourishment from the writings of other academicsrather than judges, even the sort of peripatetic judges who turn up atsymposia like this to listen, learn and inwardly digest.

What is intensely enjoyable about these sorts of confrontations is theenthusiasm with which the participants attack one another. For example,John Swan observed at the symposium:

1 K Llewellyn, The Bramble Bush: Some Lectures on Law and its Study (1930) 3(emphasis in original), recently reprinted as K Llewellyn, The Bramble Bush: The ClassicLectures on the Law and Law School (New York, Oxford University Press, 2008).

2 HJ Laski, A Grammar of Politics, 4th edn (London, Allen & Unwin Ltd, 1963) vi.



From my point of view, I can’t see what could possibly be gained by theacceptance of Stephen [Smith’s] argument that, for example, the law of damagesfor breach of contract is not part of the law of contract but should instead bedealt with in a discussion of both contract and tort damages, with tort damagesseeming to run from cases of personal injury through negligent misrepresentationto trespass, defamation and beyond.

The ‘what’s your point’ rejoinder was much favoured by most of theparticipants in both camps most of the time. While the chapters publishedhere are generally more restrained in tone than the symposium itself,together they represent significant and closely-argued contributions to ourcollective understanding of contract law. We are fortunate to have thework of both the theoreticians and the Llewellynites gathered together herein permanent form.

Justice Ian BinnieSupreme Court of Canada

28 May 2008

vi Foreword



Contents

Foreword vContributors ixIntroduction xi

1. The Limits of Contract 1STEPHEN A SMITH

2. Border Control: Some Comparative Remarks on theCartography of Obligations 25HELGE DEDEK

3. Principle in Contract Law: the Doctrine of Consideration 51STEPHEN WADDAMS

4. Contractual Interpretation at Common Law and CivilLaw: An Exercise in Comparative Legal Rhetoric 77CATHERINE VALCKE

5. Consideration and the Morality of Promising 115ANDREW S GOLD

6. Justifying Damages 139CHARLIE WEBB

7. Damages and the Right to Performance: A GoldenVictory or Not? 171ROBERT STEVENS

8. Estoppels and Rights-Creating Events: Beyond Wrongsand Promises 199ANDREW ROBERTSON

9. Lumley v Gye and the (Over?) Protection of Contracts 225G H L FRIDMAN

10. Contracting Out of Liability for Deceit, InadvertentMisrepresentation and Negligent Misstatement 237MARK P GERGEN



11. Assignments, Trusts, Property and Obligations 267ANDREW TETTENBORN

12. The Nature of Equitable Assignment andAnti-Assignment Clauses 283C H THAM

13. Coming to Terms with The Great Peace in CommonMistake 319KELVIN F K LOW

14. Contractual Mistake, Intention in Formation andVitiation: the Oxymoron of Smith v Hughes 341MINDY CHEN-WISHART

15. From Morgan to Etridge: Tracing the (Dis)Integration ofUndue Influence in the United Kingdom 379RICK BIGWOOD

Index 431

viii Contents



Contributors

RICK BIGWOOD, LLB, PHD, Professor of Law, University of Auckland

RICHARD BRONAUGH, BA, MS, PHD, BLITT, Professor of Law and Profes-sor Emeritus of Philosophy, University of Western Ontario

MINDY CHEN-WISHART, BA, LLB, LLM, MA, Reader in Contract Law,Merton College, University of Oxford

HELGE DEDEK, First and Second State Examination in Law, LLM, PHD,Assistant Professor of Law, McGill University

GERALD H L FRIDMAN QC, MA, BCL, LLM, FRSC, Professor Emeritus ofLaw, University of Western Ontario

MARK P GERGEN, BA, JD, Fondren Foundation Centennial Chair forFaculty Excellence, University of Texas

ANDREW S GOLD, BA, JD, Associate Professor of Law, DePaul UniversityCollege of Law

KELVIN F K LOW, LLB, BCL, Associate Professor of Law, The University ofHong Kong

JASON W NEYERS, BA, LLB, MST, Associate Professor of Law and CasselsBrock LLP Faculty Fellow in Contract Law, University of Western Ontario

STEPHEN G A PITEL, BA, LLB, LLM, PHD, Associate Professor of Law,University of Western Ontario

ANDREW ROBERTSON, LLB, LLM, PHD, Professor of Law, University ofMelbourne

STEPHEN A SMITH, BA, LLB, PHD, William Dawson Scholar and Professorof Law, McGill University

ROBERT STEVENS, BA, BCL, Professor of Commercial Law, UniversityCollege London

ANDREW TETTENBORN, MA, LLB, Bracton Professor of Law, University ofExeter



CHEE HO THAM, LLB, BCL, Associate Professor of Law, SingaporeManagement University

CATHERINE VALCKE, LLB, LLB, LLM, JSD, Associate Professor of Law,University of Toronto

STEPHEN WADDAMS, BA, MA, PHD, LLB, LLM, SJD, FRSC, UniversityProfessor and Goodman/Schipper Professor of Law, University of Toronto

CHARLIE WEBB, BA, LLM, PHD, Lecturer, London School of Economicsand Political Science

x Contributors



Introduction

JASON W NEYERS, RICHARD BRONAUGHAND STEPHEN G A PITEL*

The chapters in this book arose from the ‘Exploring Contract Law’symposium held at the University of Western Ontario, Faculty of Law, on10–11 January 2008. The symposium brought together leading contractscholars, private law theorists and judges from around the common lawworld. The purpose of the symposium was to have these thinkers examinecontract law with fresh eyes—to explore it anew. Thus, the presenters wereasked to explore contract law (or its related doctrines) in one of threeways. First, they could (re)explore doctrines that are considered tangentialor antiquated. This task was undertaken, for example, in the presentationsby Mindy Chen-Wishart and Gerald Fridman. Second, they could explorewhat appeared to be settled principles in light of recent case law develop-ments. This was done most clearly in the presentations by Robert Stevens,Kelvin Low and Rick Bigwood. Third, they could explore black lettercontract law from a theoretical or comparative perspective. Many of thepresentations did this in one form or another.

As noted by Justice Binnie in his foreword, collectively these explorerscan also be divided into two groups, though not without overlap. Someexplorers, particularly the authors of the chapters which open this collec-tion, took a theoretical perspective. Others took a more practical view,concerned with doctrine and its impact on practitioners and the courts.Depending on their own leanings, readers will find that some of theresulting chapters blaze new trails through terrain that feels more familiarto them, while others bushwhack into less charted territory.

Beyond the formal presentations, which have become chapters in thiscollection, the symposium was enriched by commentary and questionsfrom scholars and theorists such as Peter Benson,1 Ralph Cunnington,2

Daniel Markovits,3 John McCamus,4 Michael Pratt5 and John Swan6 and

* Each of the Faculty of Law, University of Western Ontario.1 P Benson (ed), The Theory of Contract Law: New Essays (Cambridge, Cambridge

University Press, 2001).2 R Cunnington and D Saidov (eds), Contract Damages: Domestic and International

Perspectives (Oxford, Hart Publishing, 2008).3 D Markovits, ‘Contract and Collaboration’ (2004) 113 Yale Law Journal 1417.4 JD McCamus, The Law of Contracts (Toronto, Irwin Law, 2005).5 M Pratt, ‘Promises, Contracts, and Voluntary Obligations’ (2007) 26 Law and

Philosophy 531.6 J Swan, Canadian Contract Law (Markham, LexisNexis Butterworths, 2006).

Columns Design Ltd / Job: Neyers / Division: Introduction /Pg. Position: 1 / Date: 26/2


from justices of the Supreme Court of Canada, the British Columbia Courtof Appeal and the Court of Appeal for Ontario. For two days, thoseassembled were treated to a rich diet of scholarship and informal exchangeand this collection is the result of that endeavour.

* * *

Stephen Smith’s chapter opens the book by seizing on its theme ofexploration.7 He does this not in the way of those seeking out new lands indistant places but in the way of those who ‘advance tentative observations’on what is already in place. Thus he wishes to improve the existing map ofcontract law, that is, to be a logical, local cartographer and not a scout. Tothis taxonomical end, he argues that many of the rules that are convention-ally regarded as contractual, and that, as a consequence, are regularlyinvoked to support a particular theory of contract law or explained on thebasis of such a theory, are not in fact contractual at all. Instead, theyshould be regarded as belonging to the (hitherto unrecognised) general partof the law of obligations. Smith concludes that failing to pay attention tocontract law’s special border—which he labels ‘vertical’—with the generalpart can lead ‘scholars, judges and lawyers to apply the wrong principles tounderstand legal rules, to draw the wrong inferences from those rules, andto fail to make appropriate generalisations’.8

Helge Dedek’s chapter is inspired by the works of Peter Birks in generaland Stephen Smith in particular and offers a comparative law perspectiveon the taxonomic debate in common law scholarship.9 Of course, theattempt to impose a logical order on the law is the hallmark of civilian,and particularly German, ‘legal science’ of the nineteenth century. Withoutseeking to discredit the undertaking, Dedek acknowledges that modernapproaches have surpassed nineteenth century formalism by explicitlytaking the normative (or, as Smith puts it, ‘moral’) implications of legalclassification into account. However, drawing from experiences with theGerman civil law, Dedek focuses on the downside of dividing the law ofobligations into a ‘general’ and a ‘special’ part, namely technical problemsregarding the formulation and application of abstract ‘general’ rules andthe problem of the ‘hardening of categories,’ a phenomenon which couldlead to an intellectual rigidity that curtails the argumentative potential oflegal discourse. The latter problem is illustrated by comparing howdifferent legal traditions conceptualise the ‘reliance interest’ in contractdamages.

7 ‘The Limits of Contract’ ch 1.8 Ibid, at Part VIII.9 ‘Border Control: Some Comparative Remarks on the Cartography of Obligations’ ch 2.

xii Introduction



In his chapter, Stephen Waddams examines the concept of principle—that is, what we mean by calling something a principle—in relation to thedoctrine of consideration.10 Whereas Smith and Dedek are concerned withthe general and special borders of contract law, Waddams is moreconcerned with content and substance than with form and taxonomy. Yetthis concern has formal or taxonomical effects relevant to the prior twochapters. Waddams notes that Blackstone did not think of contract as anindependently existing field of law but rather thought of it as part ofseveral different areas of the law. Hence it would have no special bordersand would be found in several different places on a map. Waddams arguesthat a historical study cannot establish either (a) the correct meaning of theword ‘principle’ or (b) the ‘correct’ rules of contract law. Being unable tofind historically a single or fixed content for the doctrine of consideration,Waddams concludes simply that the regular appeal by the courts to theconcept of principle has historical significance and shows that neither theusage nor the law (and its taxonomy) is immutable or eternal.

Catherine Valcke’s chapter compares two national legal systems and sofurther complicates the issue of the taxonomy and content of contract.11

Her aim is twofold: (a) to demonstrate that there are peculiarly differentFrench and English thought structures animating the intellectual life andlaw of French and English societies, and (b) that one of the purposes ofcomparative law, as a self-standing academic discipline, is to uncover andcompare these different thought structures. The chapter draws parallelsbetween French contract law (subjective intention as it informs issues ofmistake and interpretation) and Rousseau’s conception of the state, anddraws parallels between English contract law (objective intention as itinforms the same issues) and Hobbes’ conception of the state. Valckeargues that although these conceptions may differ in their social value, theyare internally accurate in that each correctly reflects the distinctive self-understandings of contract law in the two nations. Such a conclusion hasclear implications for exploring or ‘mapping’ contract law through formand content.

Andrew Gold examines yet another border issue for contract law,namely the line between a moral agent’s understanding of the obligation ofpromising in ordinary life and promissory obligation as enforced incontract law.12 Gold addresses the claim, recently made by Seana Shiffrin,that the doctrine of consideration presents a troubling divergence from

10 ‘Principle in Contract Law: the Doctrine of Consideration’ ch 3.11 ‘Contractual Interpretation at Common Law and Civil Law: An Exercise in Compara-

tive Legal Rhetoric’ ch 4.12 ‘Consideration and the Morality of Promising’ ch 5.

Introduction xiii



promissory morality.13 Noting that Shiffrin considers matters only from thepromisor’s point of view, Gold takes up the position of the moral promisee.Here Gold finds that promissory morality and contract law do not diverge.He affirms that when a gratuitous promise is broken, the moral promiseegains a right personally to rebuke the promisor, but no more—neither inmorality nor in law. But when the promise received by the promisee is partof an agreement grounded in an exchange then a stronger right toperformance is acquired by the promisee, both morally and legally. In fact,this emergent moral right is to a remedy from the promisor in case ofbreach and justifies creating (as substitute for self-help) the legal power forcoercion through the state—something well beyond rebuke. The existenceof contractual consideration thus serves, Gold argues, to justify an aug-mented moral right, a right that cannot exist in this way when the promisehas been gratuitous. This conclusion is contrary to Shiffrin’s divergencethesis about the consideration doctrine. The quid pro quo requirement ofthe bilateral contract should not present an embarrassment for the agentseeking to live a moral life.

Gold’s chapter makes the moral and legal right to a remedy, typicallydamages, a consequence of a promisee’s having provided consideration in abilateral contract to the promise breaker. Yet Smith, in the first chapter inthis volume, would not regard damages as part of contract law proper. Tounderstand damages, he holds, one must look up to what is more generalin law because ‘the duty to pay damages is not uniquely a response tobreach of contract’.14 Still, one cannot deny (and Smith does not deny) thatdamage rules must be understood in order to appreciate the practice ofenforcing contracts.

One turns then to the chapter by Charlie Webb, who asks what justifiesan award of damages for breach of contract when performance clearly iswhat a party contracts for.15 Webb observes that: ‘To obtain performanceis one thing; to receive a sum of money to make up for the losses caused bynot obtaining performance is something different.’16 For example, if theright to performance is taken seriously, then it would seem that specificperformance should be the natural remedy. Nonetheless, the occasion forthe performance may be past or its reclamation may demand unduesupervision by the court. What then? Webb claims that this contractualright ‘can, sometimes, be effectuated through an award … which theclaimant uses to purchase an equivalent “performance” from an alternative

13 S Shiffrin, ‘The Divergence of Contract and Promise’ (2007) 120 Harvard Law Review708.

14 ‘The Limits of Contract’ ch 1 at text following n 21.15 ‘Justifying Damages’ ch 6.16 Ibid, at text following n 10.

xiv Introduction



source’.17 The aim is to find remedies which as much as possible adhere tothe ‘special’ domain of contract law. Any damages award that goes beyondsimulating performance, Webb claims, requires reflection on ‘the normsand ideals which shape and justify the law’.18 Thus, Webb would seem tobe responsive to Smith’s distinction between ‘special’ and ‘general’ since itappears that once one leaves the realm of specific performance anddamages which simulate performance, it appears as if one must departfrom the ‘specific’ and pass through into the ‘general’ area of the law ofobligations.

In the second damages chapter,19 Robert Stevens assesses whether therecent House of Lords decision in Golden Strait Corporation v NipponYusen Kubishika Kaisha (The Golden Victory) was rightly decided.20 Hisultimate conclusion is that the result is defensible if one differentiatesbetween damages representing the right to performance (Webb’s focus) andthose awarded for consequential losses. Using examples drawn from awide range of contracts (such as sale, carriage and construction) Stevensargues that this differentiation is necessary since: (i) the rules for theassessment of these two types of damages (such as mitigation, remoteness,timing of assessment and ancillary benefits) are quite different, and (ii) it isthe only coherent way to justify otherwise inexplicable differences indamages quantification. Thus Stevens argues that although damages forbreach of contract are commonly thought to be awarded only to compen-sate the claimant for loss or, more rarely, to strip the defendant of a gain,‘general’ damages are commonly quantified by reference to the value of thecontract right which the defendant has infringed, despite the absence ofany consequential loss. Moreover, he claims that once awards of ‘substitu-tive’ damages are accepted, it is doubtful whether there is significantauthority supporting the award of a remedy assessed by reference to thedefendant’s gain.

Having considered the general border of contract law through the law ofremedies, there is the special border still to be examined. How doescontract interact with or separate itself from other areas of law on thesame plane of a legal structure? This is the place where every field of law,while ‘special,’ can be interactive with others. The next five chaptersexplore the special or horizontal border.

In Anglo-Australian law, it is unclear whether proprietary and equitableestoppel form part of the law of contract, part of the law of wrongs, or aseparate part of the law of obligations. Adopting Birks’ taxonomy thatorganises private law according to various rights-creating events, Andrew

17 Ibid, at Part VII.18 Ibid.19 ‘Damages and the Right to Performance: A Golden Victory or Not?’ ch 7.20 [2007] 2 AC 353 (HL).

Introduction xv



Robertson argues that these estoppels cannot be seen as part of the law ofwrongs because it is possible to identify a series of events which give rise toprimary rights that are recognised by the law prior to and independent ofany infringement of those rights.21 Robertson then argues that thesedoctrines cannot be seen as part of the law of contract either because apromise is neither necessary nor sufficient to establish liability. As a result,Robertson concludes that proprietary and equitable estoppel must beclassified as sui generis and therefore belong in Birks’ category of ‘other’events. They are specially (horizontally) separate from contract law andindeed not interactive.

Gerald Fridman’s chapter is about excessive crossing of the borderbetween tort and contract.22 By investigating the tort of inducing breach ofcontract, he assesses the protection that the reasonable expectations of thecontractual parties receive in the law of tort. Fridman’s argument is thatthe foundational, yet extremely controversial, decision in Lumley v Gye,23

was not supported by precedent and is a significant example of judicialactivism. He criticises the courts, old and new, for being too willing to usetort law to protect contracts. In the second half of the chapter, Fridmanargues, in opposition to the recent judgment of the High Court of Australiain Zhu v Treasurer of New South Wales,24 that the defence of justificationon moral grounds should be significantly expanded to better protect thosewho induce breaches of contracts.

In his contribution, Mark Gergen examines American law on theeffectiveness of agreements that absolve an actor from liability for mislead-ing another.25 This is to consider the interaction of contract and the law ofwrongs. Although American courts are divided on the issue, Gergencontends that the courts that refuse to enforce provisions which exculpatefrom fraud are correct. He argues that innocent parties can be adequatelyprotected from baseless accusations by rules requiring that fraud be pledwith specificity and proven by clear and convincing evidence. In relation toinadvertent misrepresentation, Gergen argues that terms that bar a con-tract claim on a representation should also bar claims in rescission,restitution, or negligent misstatement since this is a contract law issuerather than one of equity, restitution or tort. Turning to negligent misstate-ment, Gergen observes that an exculpatory agreement will determine theexistence and scope of an actor’s duty of care since invited reliance is a sinequa non of liability. In this important respect, Gergen argues that negligent

21 ‘Estoppels and Rights-Creating Events: Beyond Wrongs and Promises’ ch 8.22 ‘Lumley v Gye and the (Over?)Protection of Contracts’ ch 9.23 (1853) 2 El & Bl 216, 118 ER 749 (QB).24 (2004) 218 CLR 530 (HCA).25 ‘Contracting Out of Liability for Deceit, Inadvertent Misrepresentation and Negligent

Misstatement’ ch 10.

xvi Introduction



misstatement resembles contract, in particular the doctrines of promissoryestoppel and third beneficiary in American law, and is unlike the generaltort of negligence.

The rights under a contract can be, and frequently are, assigned to athird party. Less frequently, a contracting party can create a trust of therights under a contract in favour of a third party beneficiary. In his chapter,Andrew Tettenborn compares and contrasts these two approaches totransfer.26 The distinction is particularly important in light of two rela-tively recent English cases, Don King Productions Inc v Warren27 andBarbados Trust Co v Bank of Zambia.28 These cases take the position thatthe two approaches are different, and if one is unavailable—for examplebecause of an express prohibition on assignment—the other remains open.Tettenborn challenges these decisions as failing to reflect commercialreality and creating a distinction lacking any real difference. Tettenborn’sapproach takes a broad and strong view of clauses in a contract whichprohibit the benefits under the contract from being assigned to a thirdparty. On that view, such clauses should prohibit the creation of a trust ofthe contract’s benefits.

In contrast, Chee Ho Tham’s chapter argues for a minimalist approachto such ‘third party’ clauses.29 He supports the analysis in Don King and iscritical of Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd, acase in which the House of Lords relied on an anti-assignment clause toinvalidate a purported equitable assignment.30 In Tham’s view, equitableassignment should be understood as involving only a transfer of the rightto grant a release and its corollary, the right to decide to bring legalproceedings on the chose in action. Anti-assignment clauses thus cannotprevent an assignor from effectively transferring, in equity, his or her rightto grant such a release. Tham’s claim is that these clauses are less effectivethan the courts currently think they are.

The final group of chapters deals with the concrete doctrines of commonlaw that seek to protect parties in situations of information or powerimbalance. This time the border is not between special fields of lawhorizontally considered or as fields captured by general, vertically appreci-ated, principles of law. The issue here is about the relationship betweenpersons involved in making an agreement within the bargaining confines ofcontract formation. In his contribution,31 Kelvin Low examines the EnglishCourt of Appeal’s recent judgment in Great Peace Shipping Ltd v Tsavliris

26 ‘Assignments, Trusts, Property and Obligations’ ch 11.27 [2000] Ch 291 (CA) [Don King].28 [2007] EWCA Civ 148, [2007] 1 Lloyd’s Rep 495 (CA).29 ‘The Nature of Equitable Assignment and Anti-Assignment Clauses’ ch 12.30 [1994] 1 AC 85 (HL).31 ‘Coming to Terms with The Great Peace in Common Mistake’ ch 13.

Introduction xvii



Salvage,32 which reformulated the law of common mistake by overrulingSolle v Butcher.33 Low finds it surprising that The Great Peace has founditself under such sharp criticism—a development which demonstrates amellowing of views of both academics and judges over Denning LJ’sexcesses in Solle v Butcher. In fact, instead of criticising The Great Peace,he claims that it has wrought two principal improvements to the law: theabolition of a dual jurisdictional approach to cases of common mistakewhich caused an unnecessary amplification of uncertainty, and the clearseverance of the doctrine of common mistake from the implied conditionprecedent theory. Low contends that these improvements allow the com-mon law scope for developing its remedies of mistake beyond the bluntinstrument of declaring the contract void ab initio.

Mindy Chen-Wishart’s chapter examines the foundational and contro-versial case of Smith v Hughes34 and the objective theory of contractformation.35 She concludes that situations that are said to representexamples of subjectivity trumping the objective approach are straightfor-ward applications of objectivity when a context-specific view is taken. Infact, she argues that when this view of objectivity is applied, there is noneed, indeed no room, for a resort to subjectivity. Her chapter alsoattempts to stabilise the language of ‘mistake’, ‘defective consent’ and‘void’ and make distinctions between contract formation and vitiation. Herargument is that this distinction explains why known non-correspondenceof any term prevents contract formation, while mistaken assumptions mustbe shared and fundamental to void a contract. This distinction also allowsus to map the related areas of rectification, non est factum, mistakenidentity and misrepresentation.

In the final chapter of the book, Rick Bigwood examines the nature ofundue influence, a doctrine frequently employed by vulnerable parties toset aside contracts.36 He explains the different categories of cases, inparticular highlighting the differences between cases where the undueinfluence is established on the facts of the case and cases where the undueinfluence is rooted in the relationship between the parties. He analyses indetail the fiduciary elements on which this second category of cases hasbeen based, and is critical of the House of Lords for having lost sight ofthese important underpinnings in its more recent decisions, notablyNational Westminster Bank plc v Morgan37 and Royal Bank of Scotland

32 [2003] QB 679 (CA) [The Great Peace].33 [1950] 1 KB 671 (CA).34 (1871) LR 6 QB 597.35 ‘Contractual Mistake and Intention in Formation and Vitiation: the Oxymoron of

Smith v Hughes’ ch 14.36 ‘From Morgan to Etridge: Tracing the (Dis)Integration of Undue Influence in the United

Kingdom’ ch 15.37 [1985] AC 686 (HL).

xviii Introduction



plc v Etridge (No 2).38 In Bigwood’s view, the court’s move away fromstrict fiduciary regulation in relational undue influence cases is a retrogradestep.

* * *

Both the symposium and this book have benefited from generous supportand assistance. We are indebted to the faculty, staff and students atWestern Law, particularly Dean Holloway, Acting Dean Brown, AssociateDeans Edgar and Huscroft, and our conference coordinator, MichelleBothwell, for their enthusiastic support of the symposium. We are alsograteful for the financial support provided by Western’s Research Promo-tion Fund, which financed the operating costs of the symposium, and bythe Foundation for Legal Research, which allowed us to have the editorialassistance of two talented law students, Jean-Michel Corbeil and CarrieAnn Miller. We are also indebted to the law firm of Cohen Highley LLP forgraciously sponsoring the symposium banquet and to Justice Ian Binnie forproviding the foreword to the collection.

We hope that you will enjoy reading the chapters in this book as muchas we enjoyed hearing them presented as papers at the symposium. We areconfident that they will be helpful guides to anyone who is interested inexploring contract law.

38 [2002] 2 AC 773 (HL).

Introduction xix





1

The Limits of Contract

STEPHEN A SMITH

THE THEME OF this book brings to mind two images. The first isthat of a legal scholar who, like the explorers of old, goes in searchof new or little-understood contract law terrain. The other image is

that of a scholar who advances tentative observations or arguments aboutcontract law. In form and substance, this article is an exploration in thesecond sense. Focusing on well-known legal rules, it offers tentativeobservations and arguments about the law of contract and related areas ofprivate law. But it is the first image that provides the article’s inspiration.This article seeks to provide guideposts for those who explore (knowinglyor not) the outer reaches of the law of contract.

I. EXPLORERS AND BORDERS

Contract law explorers are sometimes content to do no more than describewhat they have found on their journeys. But most such scholars alsopursue one or the other (or both) of two further aims. The first is to applyideas and concepts drawn from the broader law of contract to the new orlittle-understood rules in the hope that these ideas and concepts will assistin understanding those rules. The second aim is to apply what they havelearned about the new or little-understood rules to the broader law in thehope that these lessons will shed light on the broader area. In practice,most scholars pursue both aims, moving back and forth between particularcontract law rules and broader contract law ideas and concepts.

Neither of these two activities can be undertaken without a preliminaryidea of the scope or terrain of the law of contract. Legal exploration, by itsnature, occurs near borders. If exploratory scholars are not careful theymay leave the territory of contract without realising it. Terminologicalconfusion is a typical consequence. But the more serious consequence ofscholars failing to notice that they have left the terrain of contract is thatthey may attempt to understand cross-border law using inappropriate

Columns Design Ltd / Job: Neyers / Division: FinalECLch01 /Pg. Position: 1 / Date: 26/2


concepts (that is, contract law concepts) or may use that cross-border lawto draw inappropriate conclusions about contract law—or may do boththese things. The main reason for paying attention to contract law’sborders is to ensure that the tasks of understanding particular rules anddrawing conclusions from those rules are done properly. To be sure,contract law’s borders are not fixed. They are the product of ongoingreflection about the nature of contracts, which is informed by exploratoryscholarship. One reason scholars explore is to test, inform and, wherenecessary, revise their views about the borders between contract and otherlegal fields. But that complex task cannot be undertaken without an idea,however preliminary, of what those borders might be and on what basisthey have been constructed.

II. HORIZONTAL AND VERTICAL BORDERS

The law of obligations is conventionally divided into sub-categoriesorganised around particular kinds of obligations. Thus, the sub-category ofcontract law is associated with the obligation to perform a contract, whilethe sub-category of unjust enrichment is associated with the obligation toreverse an unjust enrichment. The third main sub-category, the law oftorts, though not identified with a single obligation, is generally under-stood to be comprised of a number of sub-categories, each of which isassociated with a particular obligation. Thus, torts textbooks typicallydiscuss, inter alia, the law of trespass (the obligation not to commit atrespass), the law of defamation (the obligation not to defame others), thelaw of nuisance (the obligation not to interfere with others’ use of theirproperty), and the law of damages (the obligation to pay sums to thosewhom one has defamed or to those against whom one has committed atrespass, committed a nuisance, etc).

It is not clear if these obligation-specific categories are meant to exhaustobligations law (so far as I am aware the issue has never been addresseddirectly), but the absence of other recognised categories suggests that this isthe working assumption of most common law scholars. Yet if this is theassumption, it raises an immediate question: How, if at all, is the idea thatthere is such a thing as a ‘law of obligations’ reflected in this scheme? If thelaw of obligations is a meaningful category, then the various obligationsthat comprise it must have something in common. Further, if the law ofobligations is a meaningful legal category, these common elements shouldbe reflected in the law. The very concept of a law of obligations appears toassume the existence of rules that cut across or otherwise tie togetherobligation-specific categories such as contract, trespass, and so on. Itappears to assume, in other words, that the law of obligations contains notjust obligation-specific rules but also general rules.

2 Stephen A Smith



The concept of a ‘general part’ of the law of obligations, though familiarto civilian lawyers,1 has not traditionally been part of the common lawlawyer’s vocabulary. It seems clear, however, that there exist rules withinthe common law that cut across or otherwise tie together the obligation-specific categories. These rules appear to be of three main kinds. The firstgroup is comprised of rules that set common pre-conditions for differentobligations. An example is the set of rules governing the creation andexistence of ‘artificial legal persons’ such as registered corporations orcharities. These rules are applied by courts in both contractual andnon-contractual settings. Only legal persons can be bound by contracts.Equally, only legal persons—as defined by the same set of rules—can beheld responsible for injuring others. Secondly, there are rules that setcommon ‘enforcement’ conditions for different obligations. Limitationperiods, for example, are often common in this sense. Thus in the UnitedKingdom both actions founded on torts and actions founded on contractsexpire six years after the cause of action accrued. Finally, there are entirecategories of obligations that are dependent upon or derived from otherobligations. The obligation to pay damages, for example, arises if and onlyif another obligation—which may be contractual or non-contractual—hasbeen breached.

Understood in this way, the law of obligations may usefully be organisedalong the lines of a schema that until now has been applied (in thecommon law anyway) primarily within the field of criminal law. Criminallaw scholars conventionally divide the rules that constitute the criminallaw into those belonging to the ‘special’ part of criminal law and thosebelonging to the ‘general’ part.2 The former is comprised of rules thatidentify particular offences (for example, theft, murder and assault), whilethe latter includes rules that are applied generally, that is to say, to morethan one kind of specific offence (for example, the defences of duress andinsanity and the concept of mens rea). The equivalent distinction withinobligations law is between rules used to identify particular primaryobligations (for example, the obligation to perform a contract, the obliga-tion not to trespass and the obligation to tell the truth) and those that areapplied to obligations generally or at least to more than one kind ofobligation. ‘Primary’ obligations, as understood here, are obligations thatare not dependent on or otherwise derived from other obligations. The

1 This distinction is especially well known in German law and scholarship: see, eg, RZimmermann, The Law of Obligations (Oxford, Clarendon Press, 1996) 29–31; K Zweigertand H Kotz, An Introduction to Comparative Law, 3rd edn, trans T Weir (Oxford,Clarendon Press, 1998) 146–7.

2 See, eg, S Shute and AP Simister (eds), Criminal Law Theory: Doctrines of the GeneralPart (New York, Oxford University Press, 2002); RA Duff and SP Green (eds), DefiningCrimes: Essays on the Special Part of the Criminal Law (New York, Oxford University Press,2005).

The Limits of Contract 3



qualification is necessary to ensure that the law of damages—which isclearly concerned with a specific type of obligation—is distinguished fromthose obligations on which it is dependent (for example, the obligation toperform a contract, the obligation not to trespass, etc).3 Though differentin important ways from the legal personality rules and rules on limitationperiods (neither of which serve to identify a distinctive obligation), thedamages rules are appropriately regarded as general because, like thepersonality and limitation rules, they cut across or otherwise tie togethervarious specific obligations. This view of damages is defended in moredetail below. For the moment, it is sufficient to accept that obligations topay damages are dependent upon, and therefore structurally differentfrom, obligations to perform contracts, to not trespass, to not commit anuisance and so on.

The distinction between the special and the general parts of obligationslaw suggests that it may be helpful to think about the organisation ofobligations law using the civilian image of a pyramid of rules. The rules onthe top (which belong to the general part of obligations law) apply to morethan one kind of primary obligation, while the rules on the bottom (whichbelong to the various special parts of obligations law) apply to individualinstances of primary obligations. It also follows from this way of under-standing obligations law that contract law’s scope should be demarcated interms of two borders. The first, which I will call the ‘horizontal’ border,separates contract law rules from the other sets of primary obligation-specific rules that occupy the bottom of the obligations pyramid. It thusseparates contract law rules from rules identifying other obligations that,while arising from different events4 and having different contents, operateon the same plane as contractual obligations. As just noted, that plane, inbroad terms, is occupied by other primary obligations such as the obliga-tion not to trespass, the obligation not to damage another’s property, theobligation not to defame and so on. It does not include obligations thataim to cure, repair or otherwise respond to a rights-infringement, such asthe obligation to pay damages.5

3 Obligations to make restitution are intentionally not mentioned in this paragraph nordiscussed elsewhere in this article because their classification raises special, and as yetunresolved, issues. In a forthcoming essay, I argue, contrary to the conventional view, thatrestitutionary duties are structurally closer to duties to pay damages than to duties to performcontracts, to not commit trespasses, etc: S Smith, ‘Torts and Unjust Enrichments, Damagesand Restitution’ in R Chambers, P Mitchell and J Penner (eds), The Philosophical Founda-tions of Unjust Enrichment Law (Oxford, Oxford University Press, forthcoming 2009). If thisis correct, restitutionary duties belong within the general part of obligations law.

4 Here as elsewhere in this article the word ‘event’ includes status events such as comingof age or entering the jurisdiction (which then give rise to obligations to respect others’persons, property, etc). The more accurate alternative of describing obligations as arising‘from events and from status’ is grammatically awkward.

5 On the classification of restitutionary duties, see above n 3.

4 Stephen A Smith



The second border, which I will call contract’s ‘vertical’ border, sets theboundary between contract law and the general part of the law ofobligations. This border is vertical because it sets the boundary betweenrules at the top and bottom of the pyramid; specifically it distinguishescontract law rules (and other primary-obligation rules) from rules that arelocated above them in the sense that they qualify these rules or regulate theconsequences of breaching the obligations to which these rules give rise.

This article focuses on contract law’s vertical border. It addresses animbalance. Legal scholars have devoted considerable attention to contractlaw’s horizontal border. There is an extensive literature on the nature of acontract (for example, whether contracts are promises, reliance-inducingstatements, transfers or something else), on the difference between contractand other fields (for example, misrepresentation law and trusts), and onthe classification of things like estoppel, the trust, bailment, gratuitousundertakings, promises under seal and simultaneous exchanges—all ofthese are concerned with the location of contract’s horizontal border.6 Incontrast, contract law’s vertical border has received little attention. Asalready noted, the very idea that there exists a general part to the law ofobligations is almost unknown in the common law.7

This neglect has had significant consequences. Admittedly, the conclu-sion that a particular rule, say a rule on duress, is within the general, ratherthan the special, part of obligations law does not tell us anything about thecontent or purpose of the rule other than that both should be in some sensegeneral, that is, applicable to more than one kind of primary obligation. Bycontrast, the conclusion that a particular rule or set of rules are within thisor that aspect of the special part of obligations law often has immediatepractical implications. It may determine, for example, the standard of careexpected of a defendant or the scope of the defendant’s responsibility forbreaching the obligation. But the neglect of contract law’s vertical borderhas had serious consequences for our understanding of the law and, in thelong run, for the law’s development. It has led to misunderstandings aboutthe nature of contract law and about the nature of certain rules wronglyassumed to be contract law rules. It has also impeded our understanding ofother legal rules. The common law has failed to understand or even torecognise certain legal categories because their members have been wronglyassumed to lie within contract law (or tort law, etc). Rules with similarfoundations and aims have been applied and studied in isolation from oneanother.

6 I discuss some of this literature in S Smith, Contract Theory (New York, OxfordUniversity Press, 2004) chs 3, 5 and 7, and S Smith, Atiyah’s Introduction to the Law ofContract, 6th edn (New York, Oxford University Press, 2005) 28–35, 65–93.

7 A notable recent exception is P Cane, ‘The General/Special Distinction in Criminal Law,Tort Law and Legal Theory’ (2007) 26 Law and Philosophy 465.




These are strong claims, made here without examples. I make nopretense of developing or defending them in full. The article’s primary aimsare more modest: to introduce the idea of contract law’s vertical border, tosuggest an approach for locating this border and, finally, to identify a fewcategories of rules that, while commonly described as contract law rules,prima facie appear to lie outside contract’s vertical border.

III. CLASSIFICATORY CRITERIA: TWO QUESTIONS

Locating contract’s vertical border is an exercise in legal taxonomy. Likeother taxonomic exercises in law, it is undertaken with the aim of makingthe rules under examination more intelligible by showing how they are likeor unlike other legal rules. This task may be undertaken in different ways;the appropriate criteria depend on the classifier’s specific interest. Forlawyers, judges, legal scholars and others with a general, rather than aspecialised, interest in the law, the criteria that are normally applied areessentially moral criteria. For example, the familiar distinctions betweentort, contract, and unjust enrichment are based on the conviction that theselabels correspond, at least in broad outline, with morally significantdistinctions (for example, between self-imposed and externally imposedobligations, and, within the latter, between obligations not to harm andobligations to return benefits). Legal scholars who question the existenceor scope of these categories generally do so not on the basis that they rejectmoral criteria, but on the basis that the legal categories, at least ascurrently understood, fail to reflect significant moral categories. Thisapproach is appropriate for anyone with a general interest in the law, asthe law is defined in large part by the fact that it purports to create moralobligations.8 The most fundamental general question about the law iswhether it succeeds in this aim, that is to say, whether the law is justified.Trying to classify legal rules on the basis of moral distinctions helps inanswering this question.

The relevant moral distinctions will differ, however, depending on thekinds of rules that we are trying to distinguish between. Thus whenattempting to distinguish contract law from its neighbours, it makes adifference which border—the horizontal or the vertical—is under scrutiny.The point of fixing the horizontal border, recall, is to distinguish betweencontract-creating events and other kinds of primary obligation-creatingevents. There are different views as to how this distinction should bedrawn, but nearly all rely on variants (or combinations) of three broadviews about the nature of a contract, each of which locates the source of

8 See, eg, J Raz, Ethics in the Public Domain (Oxford, Clarendon Press, 1994) 210–20.

6 Stephen A Smith



contractual obligations in a different morally distinctive event: (1) con-tracts are promises (or agreements); (2) contracts are transfers (of already-existing rights); and (3) contracts are reliance-inducing statements.9 Insome cases, it is possible to conclude that a particular rule lies inside oroutside contract’s horizontal border without deciding between these views.Thus, a sufficient explanation for why the rules governing trespass to landare non-contractual is that on any plausible account of the nature of acontract-creating event (promise, transfer, reliance-inducing statement orsomething similar), the trespass rules are unrelated to that event. In othercases, however, it is necessary to be more precise about the nature of acontract-creating event in order to classify the rule. Thus, to determinewhether the rules governing executed gifts are contractual it is necessaryfirst to determine whether contracts are promises, reliance-inducing state-ments, transfers or something else. Only if contracts are transfers is itplausible to conclude that executed gifts qualify as contractual events.Similarly, to determine if some or all of the rules dealing with misrepresen-tation or with estoppel might be classified as contractual, it is necessary todetermine, inter alia, if the mere inducement of reliance (without apromise) can create or modify a contractual obligation.

In the case of contract’s vertical border, where the issue is whether therule in question belongs to the special or general part of the law, the aim isto distinguish between rules that identify the existence, meaning orconsequences that flow from a particular obligation-creating event andrules that identify the existence, meaning or consequences that flow fromobligation-creating events generally (or at least in more than one case). Therelevant moral distinction is therefore between rules that should applyspecifically to contractual obligations and those that should apply moregenerally. The importance of asking whether it makes sense, morally, forthe rule to be applied generally cannot be overstated.10 Different obliga-tions are often governed by rules that, while similar in broad outline, differin their details. These differences may be nothing more than accidentalby-products of a process in which the law is developed through adjudica-tion of individual disputes. Alternatively, they may reflect differences in theapplication, but not the substance, of a common principle. Or, finally, theymay reflect a difference in underlying foundations. Only in the last case do

9 See Smith (2004), above n 6, at ch 3.10 The importance of these questions is recognised in the literature on the parallel

distinction between the general and the special parts of the criminal law, though most scholarshave tended to emphasise one or the other question, not both. Thus, Glanville Williams inCriminal Law: The General Part (London, Stevens and Sons Ltd, 1953) holds that thedistinctive feature of rules in the general part is that they are applied to multiple offences,while Michael Moore in Placing Blame: A General Theory of the Criminal Law (Oxford,Clarendon Press, 1997) argues that their distinctive feature is found in their distinctive moralpurpose.




the differences preclude classifying the rule as general. But we cannot tellwhich one is the correct interpretation without an idea of what kinds ofrules it makes sense to suppose are general and what kinds it makes senseto suppose are specific.

Equally, however, it would be a mistake to conclude that a rule can beclassified as within the general part solely on the basis that this makesmoral sense. The law that we are trying to render more intelligible is thelaw that we actually have. Any classification of rules must take intoaccount how those rules are understood and distinguished by the peoplewho actually use them. The consideration doctrine, for example, is difficultto explain on the basis of any of the leading views about the nature of acontract. It is not a part of the concept of a promise, transfer, reliance-inducing statement, or any other plausible candidate for the basic contract-creating event. It is indisputable, however, that the consideration doctrineis part of the law of contract and not a part of the general law. The reasonis found in the positive law: the consideration doctrine, whatever itspurpose, is consistently applied to—and only to—contractual obligations.Another way of putting this point is that it is almost certain that the basiccontract-creating event is not a ‘simple’ or ‘unitary’ event such as apromise, transfer and so on. Rather, the event is something more like ‘apromise given in exchange for consideration’. We know this, inter alia,because the consideration rule is applied to contractual obligations andbecause it is not applied to other obligations.

To summarise, the answers to two questions must be taken intoaccount—no stronger term is warranted—when trying to determine if aparticular rule is within the law of contract or within the general part ofthe law of obligations: (1) is the rule applied to resolve not just contractualbut also non-contractual disputes? (2) is the rule of a kind that it makessense, morally, to suppose should be applied to obligations generally asopposed merely to contractual obligations?

IV. AN OVERVIEW

This article’s substantive argument is that a significant number of rulesdiscussed in articles and textbooks ostensibly devoted to contract law arepart of the general law rather than part of contract law. Before discussingthese rules, however, it may be useful to say a few words about what iscontractual. Briefly, and allowing that the terms are subject to varyinginterpretations, the law of contract comprises what would be labeled inmost Anglo-American textbooks as the rules dealing with offer andacceptance, consideration and intention to create legal relations; as well asthose concerned with the incorporation, implication, and interpretation ofcontract terms (the latter group to include at least some of the rules on

8 Stephen A Smith



mistake, frustration, termination and unenforceable contracts). These rulesare contractual because, on the one hand, as a matter of positive law theyare applied only to contractual disputes, and, on the other hand, because itmakes sense to regard them as contractual. It makes sense because theydefine the existence or meaning of the basic contract-creating event (forexample, a promise, transfer etc) or they qualify the existence or meaningof this (and only this) event. The rules on offer and acceptance, forexample, identify the basic contract-creating event, while the rules onconsideration and intention to create legal relations qualify the test for theexistence of that and only that event. Similarly, the rules on incorporationand interpretation of terms define the meaning of the event (for example,the meaning of a promise), while the rules on implication of terms(specifically implied-in-law terms) qualify that meaning and only thatmeaning.

Topics that are discussed in nearly every contract textbook such asdamages, specific performance, duress and undue influence are notincluded in this list. Further, in so far as the excluded topics are indeednon-contractual, the reason, in many cases, must be that they lie outsidecontract’s vertical border. The rules on damages, specific performance,duress and undue influence, for example, are regularly applied by courtswhen they are resolving contractual disputes. If these rules are notcontractual, the explanation cannot be that they are a part of misrepresen-tation law or negligence law, that is to say, that they belong to anotherspecial part of obligations law. Rather, the explanation must be that theyare within the general part of obligations law, that is to say, that they applyboth to contractual and non-contractual obligations. It is not possiblewithin this article to discuss all or even most of the rules that fit thisdescription. Instead, I will focus on three categories of rules that receivesignificant attention in nearly all contract textbooks and that are regularlyinvoked by contract scholars either to support a particular view of contractlaw or to be illuminated by that view, but which belong, I will argue, to thegeneral part of obligations law.

To avoid misunderstandings, my argument is not that the rules examinedbelow should never be discussed in contract textbooks, courses, articlesand so on. The lawyers, judges and students who are the main readers ofcontract law literature typically approach contract law questions from aproblem-solving perspective. They want to solve real-life (or hypotheticalreal-life) contract problems, and to do that they need to be familiar with allthe rules that are applied to contractual disputes, whatever their origin.Further, we can often learn important things about how courts understandcontracts by seeing how general rules are applied to contractual disputes. Iwill say more about this below. For the moment, it is sufficient to observethat while there are good reasons to discuss general rules in contracttextbooks and so on, this should be done self-consciously. There is always




the danger, as mentioned earlier, of drawing the wrong implications fromthese rules or using the wrong tools to examine them. In the case oftextbooks, at least, it is not clear to me if the issues discussed in this articleare even raised. Every author must of course make decisions about what toinclude and what to exclude. Some of the rules that are normally excludedfrom contract textbooks are part of the general law; for example, the ruleson legal personality are not usually discussed in contract textbooks. But itis not clear what, if any, criteria are employed in such decisions. The dozenor so contract textbooks on my bookshelves all contain explicit referencesto the borders between contract and tort, contract and unjust enrichmentand so on, but there is no mention of anything resembling the distinctionbetween contract law and the general law which is the focus of this article.

V. RIGHTS ARISING FROM NON-PERFORMANCE (1): THE LAW OF(ORDINARY) DAMAGES

Every common law contracts textbook discusses the rules governingdamages for breach of contract, usually in considerable detail. Oftenregarded as the most famous article in English on contract law, Fuller andPurdue’s ‘The Reliance Interest in Contract Damages’11 focuses, as the titlestates, on damages rules. Contract courses in common law jurisdictionsfrequently begin with a discussion of damages. When I studied contractlaw, the first cases I read—Hadley v Baxendale12 and Peevyhouse vGarland Coal & Mining Co13—were both about damages. But the rulesgoverning contract damages are not contract law rules. Nor—though thepoint is a fine one—do they belong elsewhere within the special part ofobligations law. Although the damages rules govern a specific type ofobligation—to pay damages—this obligation operates on a different planethan the primary obligations to performs contracts, not to trespass, not toinjure another and so on. The obligation to pay damages arises on thebreach of any primary obligation. It is therefore a part of the general law.

Damages orders come in different shapes and sizes.14 But in the typicalcase an order to pay damages compels the defendant to do either or bothof two things. The first is to pay a sum equivalent to the cost of repairing,replacing or purchasing a substitute for whatever property or service was

11 (1936) 46 Yale Law Journal 52.12 (1854) 9 Exch 341, 156 ER 145.13 382 P 2d 109 (Okl 1962).14 SM Waddams, The Law of Damages, 4th edn (Toronto, Canada Law Book, 2004); AM

Tettenborn, The Law of Damages (London, LexisNexis UK, 2003); S Smith, ‘The Law ofDamages: Rules for Courts or Rules for Citizens’ in R Cunnington and D Saidov (eds),Contract Damages: Domestic and International Perspectives (Oxford, Hart Publishing,2008).

10 Stephen A Smith



damaged, destroyed, lost, taken or not provided as a direct result of thedefendant’s breach of duty.15 The second is to pay a sum equal to anylosses that the claimant suffered indirectly as a result of the above breach,for example losses arising from the delay in receiving actual or substituteperformance. Whether these are separate duties16 or different aspects of thesame duty17 is an interesting question. For present purposes the importantpoint is that both duties are distinct from the duty to perform a contract.The latter is a duty to do what you said you would do. Althoughcontracting parties can stipulate that a particular sum must be paid in caseof breach, the duty to pay ordinary damages is not grounded in any suchagreement.18 Defendants are required to pay damages not because theyagreed to pay damages, but because they breached the contract. The dutyto pay arises at the moment of injury or termination and the content of theduty is set by various rules, including those on remoteness, mitigation andso on.19 When applying these rules it is necessary to determine the contentof the defendant’s contractual obligation; however, that obligation does notitself determine the amount of damages. The defendant’s contractualobligation is to deliver goods, construct a building, provide a service and soon. If this obligation is not met, the defendant is normally liable for thecost of obtaining a substitute. That cost is not set by the contract.Moreover, the defendant is also liable for whatever consequential losses theclaimant suffers as a result of late delivery. Again, this sum is not set by theparties’ contractual agreement. This is true even where the defendant isfully aware, when entering the contract, of the law on damages. Beingaware that one’s actions may have certain consequences does not meanthat one agrees to those consequences. Contracting parties agree only tothe consequences to which they have actually agreed. Thieves do not agreeto be punished merely because they were aware, at the time they commit-ted their crimes, of the sanctions for theft.

The conclusion that damages rules are not contractual does not rest,however, solely on the fact that the duty to pay damages does not have thesame source as ordinary contractual duties. As mentioned earlier, it is bothlogically possible and in practice almost certainly the case, that contractlaw is mixed in the sense that some contractual duties (or limits on thoseduties) derive from a promise (or reliance-inducing statement, transfer etc),

15 Where the breach consists merely of late performance, no damages are awarded underthis first heading.

16 As I suggest in S Smith, ‘Substitutionary Damages’ in R Grantham and C Rickett (eds),Justifying Remedies in Private Law (Cambridge, Cambridge University Press, 2008).

17 As argued by E Weinrib in ‘Two Models of Damages’ in Grantham and Rickett, ibid.18 Nor is payment of the stipulated sum the same as payment of damages. A court order to

pay a stipulated sum is simply an order to pay a debt; it is not an order to pay damages.19 The qualifying rules differ depending on which of the two kinds of damages just

described is at issue: see Smith, above n 16.




while others derive from a different source (for example, a general concernfor fairness in contracting). A duty to pay damages might be thought to fallwithin the latter category. Specifically, it might be argued that a duty to paydamages is an implied-in-law contractual term, that is to say, a term thatthe law implies into a contract unless the parties stipulate otherwise.20

Courts and legislatures commonly imply non-mandatory terms into con-tracts. Whether non-mandatory implied-in-law terms are properlyregarded as creating contractual as opposed to tort-like duties is a matterfor debate,21 but however they are classified they are not part of thegeneral law. Implied-in-law contractual terms impose specific duties withinspecific relationships. It might be argued, therefore, that the duty to paydamages is effectively an implied-in-law stipulated damages clause.

The difficulty with this response—and at the same time the reason thatdamages rules are properly considered part of the general, rather than thespecial, law—is that the duty to pay damages is not uniquely a response tobreach of contract. The duty to pay damages is a general duty in the sensethat it arises not just on the breach of contractual duties, but also on thebreach of other primary duties, such as duties not to trespass, cause anuisance, negligently harm others and so on. For both contractual andnon-contractual wrongs, the basic measure(s) of damages are—and shouldbe—the same. In each case, the defendant must pay for the cost of‘substitute performance’ in the form of repairing or purchasing a substitutefor whatever property or service was damaged, destroyed, lost, taken ornot provided directly as a result of the defendant’s breach and, in addition,must pay a sum equal to any consequential losses that the claimantsuffered indirectly as a result of the breach.22 This is appropriate: if the aimof the damages is to provide substitute performance and compensate lossesarising from a failure to perform an obligation, damages should inprinciple be equally available irrespective of the nature of obligation.23

In practice, the actual damages a claimant obtains may differ dependingon whether the underlying duty is contractual or non contractual. But withrare exceptions this is not because different principles are applied to

20 Some economist-lawyers explain damages in precisely this way; see eg, I Ayres and RGertner ‘Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules’ (1989)99 Yale Law Journal 87; JS Johnston, ‘Strategic Bargaining and the Economic Theory ofContract Default Rules’ (1990) 100 Yale Law Journal 615.

21 See Smith (2004), above n 6, at 307–14.22 The not uncommon view that the distinction between damages assessed at the level of

cost of cure and damages assessed at the value of performance, reveals something importantabout the nature of contractual obligations, is difficult to maintain once it is recognised thatclaims for tort damages for the loss or destruction of personal property raise similar issuesthat are answered in similar ways: see in particular Waddams, above n 14, at 3–82.

23 Subject to the qualification that the obligation in question is not an obligation to repairor compensate, as in the case of an obligation to pay damages or make restitution: see Smith,above n 3.

12 Stephen A Smith



contractual and non-contractual breaches, but rather because the applica-tion of the general principles underlying the law of damages is sensitive tothe nature of the duty that was breached. The contents of duties to pay forsubstitute performance and duties to compensate for consequential lossesare determined, at least in part, by the nature of the duty that is breached.To determine which losses were caused by a breach of duty you need toknow what that duty is. If the duty is to produce a certain outcome (as istrue of most contracts) then the direct and indirect consequences of failingto produce that outcome will be different from those associated with, say,breaching a duty to take care. It is this difference in application of the sameprinciple, rather than a difference in principles, that explains why the basicmeasure of damages obtained for breach of contract is often greater thanthat received, on similar facts, for tort claims.

The same is true of the limitations on damages imposed by the rulesgoverning remoteness and mitigation. Although the basic principles ofremoteness and mitigation of loss are the same for contract and tort claims,their application may differ because, again, those principles often requirecourts to look at the nature and scope of the primary duty. While is easierto say what ‘remoteness’ is not than what it is (for example, it is not merelyabout probability), in broad terms remoteness is about ‘responsibility for’.Assessments of responsibility for losses are based, in part, on the priorrelationship between the victim and the injurer. The relationship betweencontracting parties is typically—though not always—closer than the rela-tionship between the parties to a tort dispute. The explanation of mitiga-tion is similar. Although the term is used more commonly in contract casesthan tort cases (in tort the concept of mitigation is usually subsumedwithin the broader notion of remoteness), the underlying principle appliesequally to tort claims. In both contract and tort, claimants cannot recoverfor losses that they could have reasonably avoided.24 What counts asreasonable will often (though not always) differ depending on the nature ofthe breach. The victim of a contractual breach will often be expected tocontinue dealing with the wrongdoer.25 But this merely shows that ‘reason-able’ must be understood in context. The importance of the remotenessand mitigation rules lies in what they tell us about the law of damages, notwhat they tell us about the nature of contractual obligations.26

24 This is not to deny that courts sometimes say that there is a difference between how lossis quantified in tort and contract claims (see, eg, C Czarnikow Ltd (The Heron II) v Koufos[1969] 1 AC 350 (HL)), though it is unclear how often, if ever, such statements have actuallyaffected the final decision in a case. The common law tradition of treating damages claims asfalling into one or another distinct body of law (‘the law of damages for breach of contract,’‘the law of damages for tort’) has encouraged this kind of thinking.

25 Payzu Ltd v Saunders [1919] 2 KB 581.26 Authors who have attempted to derive propositions about the nature of contractual

obligations from the rules on remoteness or mitigation include G Gilmore, The Death of




This interpretation of damages rules does not deny that damagesdecisions often provide useful evidence for contract scholars. If damageawards normally reflect the primary duty in the ways explained above,then the amount of damages awarded in any particular case may provide aclue as to court’s view of the duty that was breached. But care must betaken when drawing conclusions about contractual obligations from caseson contract damages. We get from damages to contract through a theory ofdamages, not contract. Only if we know why damages are awarded can wesay what a particular award implies about the nature of the primary duty.This point must be stressed because the relationship between damagesawards and the duties for whose breach they are awarded is complex. Wehave already noted that an award of ‘ordinary’ or non-adjectival damagesprovide redress in at least two quite different ways. Further complexitiesarise in cases involving nominal damages, punitive damages, damages formental distress and other forms of what might be styled ‘vindicatory’damages.27 In all these cases damage awards may well tell us somethingabout the underlying primary duty, but in none of them does the awardreflect that duty in a straightforward fashion.28 The lesson is clear: to sayanything meaningful about contract law on the basis of damages awards, itis necessary first to have a theory of why damages are awarded. And toproduce such a theory, it is necessary to view damages in their entirety, thatis, to view damages for all breaches of primary duties together. The law ofcontract damages is one part of the general law of damages.

VI. RIGHTS ARISING FROM NON-PERFORMANCE (2): THE LAW OFCOURT-ORDERED RIGHTS

Contractual obligations are private obligations in the sense that they areowed to other citizens or to entities that are treated in law as ordinarycitizens. Governments make contracts, but in so far as these contracts aregoverned by ordinary contract law rules (as they normally are) thegovernment is treated as a natural person. It follows that the rulescomprising the law of contract are directed fundamentally at citizens:contract law rules tell citizens how they should behave when they interactwith other citizens, at least within a certain sphere of activity. But a not

Contract (Columbus, Ohio State University Press, 1974) 49–53; R Danzig, ‘Hadley vBaxendale: A Study in the Industrialization of the Law’ (1975) 4 Journal of Legal Studies249; W Bishop, ‘The Contract–Tort Boundary and the Economics of Insurance’ (1983) 12Journal of Legal Studies 241.

27 R Stevens, Torts and Rights (Oxford, Oxford University Press, 2007) 59–92; D Pearceand R Halson, ‘Damages for Breach of Contract: Compensation, Restitution, and Vindica-tion’ (2008) 28 OJLS 73.

28 Stevens, ibid, at 59–92; Tettenborn, above n 14, at 25–51; Smith, above n 14.

14 Stephen A Smith



insignificant number of rules found in standard contract textbooks aredirected at courts, not citizens. In particular, the duty-imposing rulesdealing with contract remedies, save those concerned with quantifying anordinary damages award (see above), are directed at courts. These rules tellcourts how they should behave when making court orders. It is arguablethat these rules do not belong to obligations law (or even private law), letalone the law of contract. But if they are a part of obligations law, they arewithin the general part.

A. Orders to Perform an Existing Duty

Court orders fall into two broad categories.29 In the first category, theorder (merely) confirms an existing duty, as for example when a courtorders specific performance of a contract or when it orders a defendant topay damages for consequential losses. Such orders confirm the defendant’salready-existing duty (in the examples just given to perform the contract orto compensate for consequential losses). The content of orders in this firstcategory is therefore determined by the rules that govern the content of therelevant pre-existing duty. In some cases, these are contract law rules (aswhere specific performance is ordered) and in some cases they are not (aswhere damages are ordered), but they are all private law rules, directed atcitizens. Thus, the content of an order to perform a contract is determinedby the ordinary contract law rules on the interpretation, incorporation,and implication of contract terms, while the content of an order to paycompensation for consequential losses is governed by the damages lawrules dealing with remoteness of loss, mitigation and so on.

The availability of specific performance and other orders in this firstcategory is not, however, determined by the rules governing contracts,damages, or any other private law duty. The conclusion that John has acontractual duty to deliver goods to Ann tells us what John should do(deliver goods to Ann), but it tells us nothing about what the courts or anyother organ of the state should do if John fails to perform that duty. Theseare different questions. When a claimant asks a court to make an orderagainst the defendant, the claimant is asking the state to involve itself inwhat has been, until then, a private matter. There may be good reasons forcourts to accede to such requests; this explains why many court orders areavailable ‘as of right’. But such rights are rights against courts, not againstthe defendant or any other individual. The rules that govern such rights arenot a part of the law of contract. Indeed, it is not clear that they are a part

29 R Zakrzewski, Remedies Reclassified (New York, Oxford University Press, 2005) ch 5;Smith, above n 14.




of the law of obligations or even of private law.30 Although rights to courtorders arise from infringements of private rights and although the courtorders to which they give rise are themselves a source of private obliga-tions, rights to court order are strictly against the state and, as such,arguably a part of public law.

For present purposes, the final classification of rights to court orders canawait another day. What is important is to see that whether they arepublic, private, or mixed public/private, these are general rights in thesense that they are applied to disputes arising from different kinds ofobligations. A court’s willingness to order a defendant to perform anexisting duty depends in every case on the content of that duty, but it doesnot depend on whether the duty is contractual or not.31 Existing duties topay money, for example, are invariably enforced directly if the duties aredue and the claimant so requests, provided that the relevant limitationperiod has not expired. This is true whether the underlying duty is to paydamages, to pay a contractual debt or to return money paid by mistake. Inall these cases, an order to pay the money is available as of right. Bycontrast, most non-monetary orders are subject to a complex set ofpre-conditions, particularly if the relevant duty is a duty to do somethingas opposed to a duty not to do something. Once again, this is true whetherthe original duty is contractual or not. Thus courts are generally willing toenforce negative duties not to trespass, not to create a nuisance or not tobreach a restrictive contractual covenant. Conversely, positive duties,whether in contract or not, are typically enforced only if they are relativelysimple and if substitute performance is unavailable.

The generality of the rules governing specific relief is particularlyimportant for contract scholars in light of the considerable attention suchrules have traditionally received in (ostensibly) contract law literature.Probably the most famous sentence ever written in English about contrac-tual obligations—Holmes’s ‘The duty to keep a contract … means … thatyou must pay damages if you do not keep it,—and nothing else’32—isabout the rules governing specific performance. The same is true of what isprobably the best-known idea in contract scholarship of the last half-century, the theory of ‘efficient breach’.33 Contemporary contracts scholar-ship continues to devote considerable attention to remedial rules and, inparticular, to the rules governing specific performance.34 As was true of

30 See Smith, above n 14.31 See Smith, above n 16.32 ‘The Path of the Law’ (1897) 10 Harvard Law Review 457, 462.33 RL Birmingham, ‘Breach of Contract, Damage Measures, and Economic Efficiency’

(1970) 24 Rutgers Law Review 273; JH Barton, ‘The Economic Basis of Damages for Breachof Contract’ (1972) 1 Journal of Legal Studies 277.

34 See, eg, D Kimel, From Promise to Contract: Towards a Liberal Theory of Contract(Portland, Hart Publishing, 2003); SV Shiffrin, ‘The Divergence of Contract and Promise’

16 Stephen A Smith



damages rules, it may be that we can learn something indirectly aboutcontractual obligations by observing how specific performance rules areapplied. But even more than in the case of damages rules, the rules onspecific performance have a complex relationship to the concept of acontractual obligation. The rules on specific performance stipulate varioussituations in which a state actor (a court) has the authority to do or not dosomething. As mentioned already, in so far as claimants have a right tospecific performance, it is a right against the state. It is to be expected,therefore, that the rules governing the availability of specific performance(and of other court orders) will reflect concerns that are particular to thestate and to the order that the state is contemplating making. The questionthe court must consider is not whether the defendant had a duty to dowhat the claimant alleges (that question has already been answered), butwhether the court should respond to the failure to perform that duty byinvoking its power to order specific performance or, instead, by doingsomething different. It would not be surprising to find, therefore, thatinstitutional concerns (broadly defined) explain much of the law in thisarea. In practice, this is exactly what we do find. Without going into detail,the limits on specific performance (and other requests to enforce existingduties) appear to reflect, inter alia, an institutional concern for simple,easily enforced orders, and, in addition, for avoiding orders that might beregarded as involving something akin to servitude. The rules on specificperformance therefore tell us a great deal about how courts understandtheir duties, but very little about how courts understand citizens’ duties.

B. Orders to Perform a New Duty

The other category of court orders is comprised of orders that create, bythe order itself, new duties. An example is an order to pay punitivedamages. Prior to coming to court a defendant is not under a duty to paypunitive damages to the claimant. You cannot punish yourself. The duty topay punitive damages is created by the order. It is difficult to say withcertainty how many other court orders fall into this second category. Themost likely candidates appear to be nominal damages, ordinary damages incases where performance by the defendant was possible and desired whenthe parties came to court (that is, where the contractual obligationremained in force until the order), and perhaps damages for mental distressand loss of satisfaction.35 For present purposes, the important point is that

(2007) 120 Harvard Law Review 708; A Bagchi, ‘Contract v Promise’ <http://ssrn.com/abstract=1012150> (2007) (last accessed: 1 November 2008); D Markovits, ‘Making andKeeping Contracts’ (2006) 92 Virginia Law Review 1325.

35 See Smith, above n 14.




whatever the size of this category, the orders that fall within it provide verylimited and indirect evidence of the nature of the underlying obligation. Aswas true of ‘duty-confirming’ orders, the availability of ‘duty-creating’orders is not determined by contract law principles. It is determined by thegeneral principles that govern when (if ever) it is appropriate for courts tocreate new duties through court orders. But the distance from the underly-ing duties is even greater than in the case of duty-confirming orders.Duty-creating orders neither confirm the underlying primary duty (ashappens with specific performance orders) nor, with one exception, reflectthat duty (as is true of ordinary damages). A court’s decision to awardpunitive or nominal damages, for example, tells us that the court believes aduty was breached, and, in the former case anyway, may tell us somethingabout the court’s view of the seriousness of the breach and the defendant’smotives, but the content of the award is only loosely connected to thecontent of the underlying duty. Of the orders mentioned above, only ordersto pay damages that are made in lieu of ordering specific performanceappear to reflect the content of the underlying duty. In such cases, damagesare calculated on the same basis as ordinary damages.

VII. INCAPACITY, DURESS, UNDUE-INFLUENCE ANDMISREPRESENTATION: THE LAW OF RESPONSIBILITY

Courts are generally uninterested in the attributes of contracting parties,the circumstances under which they make their contracts or their reasonsfor contracting. The same rules of offer and acceptance, interpretation, andso on are applied to the old and the young, the rich and the poor and theinformed and the uninformed. But there are exceptions. In certain circum-stances, an arrangement that would otherwise create a binding contractwill not have that effect because of the contracting parties’ attributes,circumstances or reasons for entering the contract. Some of the rules thatprovide for such a result are plausibly explained as either implied-in-fact orimplied-in-law contractual terms, and, as such, as belonging to the specialpart of obligations law, if not contract law itself. This seems the mostplausible interpretation of (most of) the rules regarding mistake,36 frustra-tion and termination for breach.37 But there are many ‘defences’ thatcannot be explained in this way, in particular those provided for by therules on incapacity, duress, undue influence and misrepresentation.

The law dealing with incapacity, duress, undue influence and misrepre-sentation is large and complex. In broad terms, however, there appear to be

36 Excepting mistake as to ‘terms’, which is part of the law of offer and acceptance: seeSmith (2004), above n 6, at 366–7.

37 See ibid, at 365–74.

18 Stephen A Smith



two main ideas underlying these rules.38 The first is that individuals shouldnot be bound to contracts if their ability to consent in a meaningful waywas significantly impaired, either generally or with respect to the specificcontract in question. Most cases in which incapacity or undue influence ispleaded successfully seem best explained on this basis. The other idea thatappears to underlie much of the law in this area is that wrongdoers shouldnot profit from their own wrongdoing (where ‘profit’ includes obtainingcontractual rights). The rules that allow a contract to be set aside for anegligent or intentional misrepresentation seem best explained on thisground. A large number of rules, in particular the rules on duress, appearto be explicable on either of the above grounds. The grounds are, however,distinct. A contract may be set aside for incapacity regardless of the faultor even knowledge of the party seeking enforcement. This is a purely‘consent-based’ explanation. Conversely, a contract entered on the basis ofa misrepresentation or an unlawful pressure emanating from a third party(as opposed to the other contracting party) generally cannot be set aside. Ifall the courts cared about was consent, the source of the mistake orpressure would be irrelevant.

Most contract scholars appear to regard the law dealing with incapacity,duress, undue influence and misrepresentation as core parts of the law ofcontract. With the occasional exception of incapacity, these topics arediscussed in every contract law textbook, and they are introduced, so far asI am aware, without qualification or explanation. Some of the best-known‘contract law’ articles focus on the law of duress.39 There is a largeliterature devoted to the distributive consequences of contract law, nearlyall of which focuses on duress, mistake and related doctrines.40 Theassumption underlying these articles is that in so far as duress, etc arebased on distributive principles, contract law is based on distributiveprinciples. More generally, nearly every discussion of what is widelyassumed to be the core or most important contract law principle—theprinciple of freedom of contract—focuses on the above doctrines.41 Thusthe rules on capacity are presented as the law’s attempt to ensure thatparties have the general ability to make free choices, while the rules on

38 See ibid, at ch 9.39 RE Barnett, ‘A Consent Theory of Contract’ (1986) 86 Columbia Law Review 269; JP

Dawson, ‘Economic Duress—An Essay in Perspective’ (1947) 45 Michigan Law Review 253;J Dalzell, ‘Duress by Economic Pressure’ (1942) 20 North Carolina Law Review 341; RLHale, ‘Coercion and Distribution in a Supposedly Non-Coercive State’ (1923) 38 PoliticalScience Quarterly 470; AT Kronman, ‘Contract Law and Distributive Justice’ (1980) 89 YaleLaw Journal 472.

40 In addition to Dawson, Dalzell, Hale and Kronman, ibid, see, eg, PS Atiyah, ‘EconomicDuress and the Overborne Will’ (1982) 98 LQR 197; J Gordley, ‘Equality in Exchange’(1981) 69 California Law Review 1587.

41 See, eg, M Trebilcock, The Limits of Freedom of Contract (Cambridge, HarvardUniversity Press, 1993).




duress, undue influence and misrepresentation are presented as the law’sattempt to ensure that parties have the ability to make free choices inparticular cases.42 Nonetheless, I suggest that these rules are not contractlaw rules, but belong instead to the general part of the law of obligations.

As already noted, contract scholars typically describe the basic event thatgives rise to a contract as either a promise, a reliance-inducing statement, atransfer (of rights), or something similar. It is not a part of the ordinarydescription of any of these events that the participants have capacity, actfree from duress, etc. A promise, for example, is created (merely) byintentionally communicating an intention to undertake an obligation.43

Minors who understand the concept of a promise, as many do, areperfectly capable of making promises and most of them do so regularly. Itis also perfectly possible to make a promise while under duress or whilesubject to undue influence or a misrepresentation. This is confirmed byordinary language: we talk of ‘promises made under duress’, (not ‘non-promises made under duress’). The distinction between the concepts ofpromise and duress also helps to explain why promises made under duressare sometimes binding. Every parent is familiar with the process ofextracting promises from their children which are made under conditionsthat can only be described as duress. Yet the parents not only considerthese statements to be promises, they consider them binding promises. Thefact that Johnny was not allowed out of his room until he promised toreturn some toys to his brother does not excuse Johnny from keeping hispromise. Every treaty signed by a surrendering army is signed underduress. In the common law, contracts that are made under duress are ingeneral perfectly valid if the duress emanated from a third party. The sameis true, in broad outline, of promises made while subject to another’sinfluence or promises made as a result of a misrepresentation. This doesnot mean that promises made under duress, etc. should always or evenoften be legally binding. The point is merely that in so far as such promisesare not binding, the reason is not that they are not promises but that, in thecircumstances, the normal obligatory force of a promise is overridden orexcused.

What is true of promises is equally true of reliance-inducing statements.It is clearly possible to induce someone to rely on your statements even ifthose statements are made under duress. The hostage who phones relativesto ask them to give the hostage-taker a sum of money is usually hopingthat these relatives will rely on the call and in many cases they do just that.

42 See, eg, C Fried, Contract as Promise (Cambridge, Harvard University Press, 1981).43 J Raz, ‘Voluntary Obligations and Normative Powers’ (1972) Proceedings of the

Aristotelian Society Suppl 46; J Raz, ‘Promises and Obligations’ in PMS Hacker and J Raz(eds), Law, Morality, and Society (Oxford, Clarendon Press, 1977); Smith (2004), above n 6,at 57.

20 Stephen A Smith



Few people would hold the hostage victim morally responsible for inducingsuch reliance, but this is not because there was nothing to fit the definitionof reliance or because the reliance was unreasonable or unforeseeable bythe one who asked for it. The question whether a transfer was made issimilarly distinct from the question whether the transferor acted underduress, etc. A transfer happens whenever an individual intentionallytransfers ownership to another. One can act intentionally even when onedoes not act freely. This is why transfers made under the threat of a severesanction are sometimes perfectly valid. Many people pay taxes onlybecause they are afraid of criminal charges, yet their payments are validlegally nonetheless. When a defendant transfers money to a claimant incompliance with a court order the transfer is valid even though thedefendant had no real choice and the claimant was fully aware of this.

Admittedly, it remains possible, in theory anyway, that the basiccontract-creating event is not merely a promise, reliance-inducing state-ment, transfer and so on, but something like a ‘promise made by someonewith capacity not acting under duress or subject to undue influence or amisrepresentation’. I noted earlier that at least some of the reasons for notenforcing a promise and so on cannot be explained by extrapolation fromthe idea of a promise, transfer and so on. The rule that promises made indomestic settings are assumed not to be legally binding is an example. Suchpromises are promises in the full sense, yet the rules regarding domesticpromises are clearly a part of contract law. The law on duress, etc, mightbe explained in a similar way. For example, we might say that just as onlynon-domestic promises can create contracts, only ‘freely made’ promises,etc can create contracts.

The analogy breaks down, however, for two reasons. The first is that itmakes sense—moral sense—to suppose that the rules on duress, etc aregeneral rules. In broad terms, these rules express a principle of responsibil-ity. The idea that one should not be held to contracts to which one did notconsent reflects a notion of negative responsibility (‘I am not morallyresponsible for my actions’), while the idea that one should not be able toenforce rights obtained through one’s wrongdoing reflects a notion ofpositive responsibility (‘I am responsible for your actions’). Responsibilityin either sense is a complex notion.44 Responsibility for a promise isdifferent from responsibility for a transfer and each is different fromresponsibility for a wrong. But it seems highly unlikely that ‘responsibilityfor’ is relevant only for certain kinds of obligations and juridical acts orthat the fundamental principles differ as between different obligations. Theconcept of responsibility is general—and so is outside the vertical border ofcontract law. This way of understanding responsibility has long been

44 P Cane, Responsibility in Law and Morality (Portland, Hart Publishing, 2002).




adopted in the criminal law, where the distinction between the general andthe special parts of the law is often explained primarily in terms of thedistinction between responsibility and culpability. Not coincidentally, thevarious defences that make up the bulk of the general part of the criminallaw (for example, duress, incapacity) are in kind (though not usually indegree) similar to the contract law defences of duress, incapacity, and soon.

The other reason the analogy between ‘domestic promises’ and ‘promisesmade under duress’ breaks down is that while the rules governing domesticpromises are uniquely applied to contractual obligations, the rules onduress, etc are applied in non-contractual settings. Like the rules ondamages and the rules on court orders, they are governed by general (thatis, not specifically contractual) principles and they apply to a range ofjuridical acts. A promise to pay $100 that is made under duress (or undueinfluence, misrepresentation, etc) is not binding. Even assuming it is madein exchange for consideration and in a commercial setting and so on, itdoes not have the normal legal effect of a promise to pay $100. This isbasic first year ‘contract’ law. Yet exactly the same conclusion is reached incases where, instead of promising to pay $100, the $100 is transferredunder duress by way of a gift, a trust or a bequest. Executed gifts, trusts,and wills are not contracts. Yet just as in the case of a (potential) contract,the normal legal effects of these events is in every case denied if the personinitiating the transaction acted under duress. This is why the textbooks onunjust enrichment discuss incapacity, duress, undue influence and misrep-resentation in detail. An unjust enrichment can arise because a contractwas made under duress, but, equally, it can arise because a non-contractualjuridical act (for example, a gift, a will, a trust) was made under duress.There is ancient authority that duress is no defence to an action in tort,45

but few suppose the authority would be accepted today46 (though theredoes not appear to be a modern decision on point). In any event, a defencebased on duress can always be restyled as a defence based on necessity,which is accepted.

Admittedly, duress and other general defences do not operate identicallyacross contractual and non-contractual settings. In part, the explanationfor the differences is precisely that the relevant rules have traditionally notbeen understood as part of the general law. Contract cases in which thesedefences are raised are discussed and analysed separately from tort andunjust enrichment cases. This had led to inconsistencies. Indeed, there areinconsistencies in how these defences are applied within the law ofcontract. There has long been a recognised defence of presumed undue

45 Gilbert v Stone (1647) Aleyn 35.46 WVH Rogers, Winfield & Jolowicz on Tort, 16th edn (London, Sweet & Maxwell,

2002) 876.

22 Stephen A Smith



influence, but the courts have yet to recognise a defence of presumedduress, misrepresentation or incapacity. Fortunately, some scholars arebeginning to examine these defences from a general perspective. Scholarsworking in the area of unjust enrichment law, in particular, have long beenaware that concepts such as duress, undue influence and so on areimportant for both contract and unjust enrichment cases. More recently,scholars working on what might be called, roughly, the philosophicalfoundations of private law, have examined these concepts in generalterms.47 Consistent with the argument of this article, these scholars areattempting to do what criminal law scholars did for their subject long ago,namely, to present and analyse defences in general terms, abstracting fromthe particular offence or cause of action.

The main reason that the rules on duress etc appear to operatedifferently depending on the nature of the primary duty in question is,however, the same reason that the rules on damages and the rules on courtorders appear to operate differently. The application of these rules toparticular cases depends on the specific context, and the context ofcontract cases is typically (though not always) different from the context ofunjust enrichment or tort cases. Just as the rules on remoteness cannot bereduced to a statistical test, the rules on duress, etc cannot be reduced to amechanical test. As already noted, the defences of duress, incapacity and soon (like the rules of remoteness) reflect principles of responsibility. Respon-sibility is a complex, context-specific notion.48 Responsibility for havingmade a promise is different from responsibility for a transfer and each isdifferent from responsibility for a wrong. But the concept itself is general;it applies to contractual obligations as well as in other contexts. Anyattempt to understand the rules that reflect this principle of responsibilitymust have regard to the full range of situations in which the principle is atissue.

VIII. CONCLUSION

In criminal law scholarship, the distinction between the special and thegeneral parts of the law is well known. It has fostered the development of asophisticated general criminal law of excuses, justifications and so on. Theparallel distinction is almost unknown in the law of obligations, at least incommon law jurisdictions. Part of the explanation for this situation is nodoubt the usual lack of attention that, historically, the common law has

47 Cane, above n 7; W Lucy, Philosophy of Private Law (New York, Oxford UniversityPress, 2007) 47–144; P Birks, ‘The Concept of a Civil Wrong’ in D Owen (ed), PhilosophicalFoundations of Tort Law (Oxford, Clarendon Press, 1995) 29.

48 Cane, ibid.




paid to classificatory undertakings generally. A more explicit reason is thatthe practical consequences of neglecting to distinguish between the generaland the specific are less evident and certainly less immediate than thosearising from neglecting the distinction between the different special parts ofobligations law. But there are consequences. Failing to pay attention tocontract law’s vertical border can lead scholars, judges and lawyers toapply the wrong principles to understand legal rules, to draw the wronginferences from those rules, and to fail to make appropriate generalisa-tions. The frequent attempts by contract law scholars to draw conclusionsabout the nature of contracts from the rules governing damages andspecific performance is just the most obvious example of where this hashappened.

This article has barely scratched the surface. There is a significantamount of law that was not discussed (for example, formalities, uncon-scionability, unenforceable contracts) and what law was discussed wasexamined only in the briefest outline. The article’s substantive conclusionsare only provisional. The article’s primary aim, however, was not topronounce final judgment on the limits of contract. Rather the aim hasbeen to introduce the idea that contract law has limits, to explain whythese limits are important, and to give some idea of how they should beidentified.

24 Stephen A Smith



2

Border Control: Some ComparativeRemarks on the Cartography of

Obligations

HELGE DEDEK*

I. INTRODUCTION

A. Mapping Law’s Empire

LEGAL HISTORY HAS witnessed many attempts to impose ameasure of order onto the unpruned growth of the common law,whether it be through codification, restatement or scholarly treatment.

It has particularly been the ambition of jurists, at least since William Black-stone’s time, to provide some guidance to those who, taking recourse to thelaw, are bound to get lost in its wilderness. Legal scholars strove for explora-tion and intellectual conquest. ‘Mapping’ is one of the most frequentlyrecurring metaphors for describing those scholars’ endeavour—mirroring theenterprise of their counterparts, adventurers and scientists exploring1 andmeasuring2 the real world instead of law’s virtual reality.3 Underlying theseprojects is what Pierre Schlag has called the ‘legal aesthetic of the grid’, which

* I would like to thank Richard Bronaugh, Nicholas Kasirer, Stephen Smith andStephen Waddams for their support and helpful comments. I also would like to thankJason Neyers for his kind invitation.

1 See, for an introduction, D Arnold, The Age of Discovery, 2nd edn (London,Routledge, 2002) passim. See also DJ Boorstin, The Discoverers (New York, Random House,1983), and P Whitfield, New found lands: Maps in the History of Exploration (London,Routledge, 1998), who presents the intellectual dimension of (real-world) explorationthrough the history of (actual) maps and the science of cartography.

2 For a colourful literary account of the scientific spirit of appropriating the unknown byexploring and measuring see D Kehlmann, Measuring the World, trans CB Janeway (London,Quercus, 2007), who focuses on the travels of obsessive-compulsive nineteenth centuryexplorer–scholar Alexander von Humboldt.

3 S Waddams, Dimensions of Private Law (Cambridge, Cambridge University Press,2003) 1ff; G Samuel, ‘Can the Common Law Be Mapped?’ (2005) 55 University of TorontoLaw Journal 271. For other ‘legal science’ metaphors invoking taxonomic disciplines such as



frames ‘the law as a field, a territory, a two-dimensional space that can bemapped and charted’.4 Legal concepts are assigned a spot on the map, so that,eventually, everything, as Christopher Columbus Langdell wrote, ‘should befound in its proper place, and nowhere else’.5

A debate has been revolving for centuries around the merits of ‘mapping’the law, with voices of criticism particularly and persistently emanatingfrom the bench.6 Its most recent revival, the ensuing of a new ‘taxonomydebate’,7 is, of course, owed to the works of the late Peter Birks of Oxford.Birks, in what is probably the most well-known articulation of hisambitions, had once more adopted the geographical imagery and themetaphor of the map. After all these years, writes Birks, lawyers are stilllost and fall victim to what he calls a ‘stovepipe mentality’—knowing ‘theirlaw only in the way that many people know London, as pools ofunconnected light into which to emerge from a limited number of friendlytube stations’.8 The reason for that mentality is, according to Birks, that‘nobody has shown them the map’.9

What is new, what is so remarkable about this last attempt at ‘mapping’?Maybe, and tentatively, a twofold answer can be given to this question:first, because it is so openly and courageously anachronistic. At the turn ofthe century, after the worldwide impact of Legal Realism and the state-ments of many scholars who wrote off the ‘mapping’ enterprise as aremnant of nineteenth century legal thought,10 this move has come ratherunexpectedly; and it is not surprising that it originated in England where

anatomy, see, eg, AJ Rodenbeck, The Anatomy of the Law, A Logical Presentation of theParts of the Body of the Law (Boston, Little, Brown & Co, 1925) and the caustic review by EMunguia Jr, (1925–26) 14 California Law Review 150–52. On the role of metaphors forchange and progress in scholarship in general see JG Daugman, ‘Brain Metaphors and BrainTheory’ in W Bechtel et al (eds), Philosophy and the Neurosciences (Oxford, Blackwell, 2001)23–5; TS Kuhn, ‘Metaphor in Science’ in A Ortony (ed), Metaphor and Thought (Cambridge,Cambridge University Press, 1993) 409–19. See also N Kasirer, ‘Bijuralism in Law’s Empireand in Law’s Cosmos’ (2002) 52 Journal of Legal Education 29, 32, 34 et passim, whodeliberately uses the imagery of the ‘map’ to characterise a jurisdictional, territorial way ofthinking about law as opposed to a more abstract, ‘borderless’ approach.

4 P Schlag, ‘The Aesthetics of American Law’ (2002) 115 Harvard Law Review 1047,1055.

5 CC Langdell, A Selection of Cases on the Law of Contracts, 2nd edn (Boston, Little,Brown & Co, 1879) ix.

6 See, eg, the more recent statements in Read v J Lyons & Co [1947] AC 156 (HL) 175;Woodar Investment Development Ltd v Wimpey Construction UK Ltd [1980] 1 WLR 277(HL) 283; Attorney General v Blake [2001] 1 AC 268 (HL) 284 [Blake].

7 G Samuel, ‘English Private Law: Old and New Thinking in the Taxonomy Debate’(2004) 24 OJLS 335.

8 P Birks (ed), English Private Law, 1st edn (Oxford, Oxford University Press) vol 1,xxxv.

9 Ibid, at xxxvi (emphasis added).10 Just see Schlag, above n 4, at 1055ff. See also D Kennedy, ‘Toward an Historical

Understanding of Legal Consciousness: The Case of Classical Legal Thought in America,1850–1940’ (1980) 3 Research in Law & Sociology 3.

26 Helge Dedek



the ideological impact of the Realists (and the following deconstructivemovements that built on their legacy) was never as strong as in the NewWorld.11 However, what seems more, and, in the true sense of the word,anachronistic, is that Birks, a Romanist, openly takes recourse to, asAndrew Burrows puts it, ‘his beloved Gaius’,12 namely, the institutionalscheme ascribed to classical Roman jurist Gaius, who lived and taught lawin the second century AD.13 Gaius’s Institutes were to Birks the paradigmof scholarship successfully ‘providing a map’; he sees the English lawyers’lack of orientation as a result of the lack of exposure to the Institutes inmodern legal education.14 This inspiration for the classificatory schemeemployed by Birks, and Birks’s openness about the source of his inspira-tion, leads us to the possible second part of our answer: the Birksian callfor taxonomy is also remarkable because it seems so blatantly civilian.

Civilian scholarship has pushed its obsession with taxonomy and classi-fication much further than the common law tradition; the ‘formal ration-ality’ (Max Weber) of the civil law is one of the stereotypes that emergewhenever attempts are made to describe the differences between thecommon and the civil law.15 Particularly to those who see legal discourseas a unique and peculiar expression of ‘legal culture’, imposing a civilianclassificatory scheme—the most famous civilian classificatory scheme—onto the common law must appear to be a problematic project due to theincompatibility of ‘mentalities’.16 Bringing the English common law closerto the Continent by imbuing it with a civilian spirit of order, the Birksianproject has political implications as well, particularly against the backdropof the ongoing efforts to create a ‘European Private Law’ employing thecivilian method of codification.17

The merits of taxonomy and classification in law have, of course, beendiscussed by many on a more abstract and general level.18 A ‘classical’

11 See, eg, N Duxbury, ‘English Jurisprudence between Austin and Hart’ (2005) 91Virginia Law Review 1.

12 A Burrows (ed), English Private Law, 2nd edn (Oxford, Oxford University Press,2007) xxxi.

13 See, eg, HLW Nelson, Überlieferung, Aufbau und Stil von Gai Institutiones (Leiden,Brill, 1981) passim. On ‘Gaianism’ see also DR Kelley, ‘Gaius Noster: Substructures ofWestern Social Thought’ (1979) 84 The American Historical Review 619ff.

14 P Birks, ‘Definition and Division, A Meditation on Institutes 3.13’ in P Birks (ed), TheClassification of Obligations (Oxford, Clarendon Press, 1997) 1ff.

15 See, eg, JH Merryman, The Civil Law Tradition, 2nd edn (Stanford, StanfordUniversity Press, 1984) 90ff.

16 See P Legrand, ‘European Legal Systems Are Not Converging’ (1996) 45 Intellectionaland Commercial Law Quarterly 52, 60ff; P Legrand, ‘How to Compare Now?’ (1996) 16Legal Studies 232, 237ff.

17 Which, in itself, gives rise to hostility from the proponents of ‘law & culture’: PLegrand, ‘Antivonbar’ (2006) 1 Journal of Comparative Law 13.

18 See, eg, Schlag, above n 4. On the impossibility of ‘legal science’—likening it to thepseudo-science of ‘phrenology’—see P Schlag, ‘Law and Phrenology’ (1997) 110 HarvardLaw Review 877.

The Cartography of Obligations 27



treatment of the subject is Roscoe Pound’s article on ‘classification of law’,disenchanting the idea that a ‘formal logic’ might operate in legal taxono-mies.19 Stephen Waddams has also expressed scepticism towards the‘concept of legal mapping’ in his study The Dimensions of Private Law,drawing on an in-depth analysis of common law reasoning.20 The works ofGeoffrey Samuel stand out as a direct response to the works of Birks,critically scrutinising, for example, the value of Birks’s statement that‘Gaius was the Darwin of the law’.21

However, it is not my intention, in this venue, to partake in the generaltheoretical discussion around the raison d’être of legal taxonomies. This isnot the place for an inquiry that would involve philosophical, psychologi-cal, anthropological, sociological and linguistic questions. Rather, I want toapproach legal taxonomy from a perspective that might be called internal,looking at some of the effects and ramifications of legal categorisationwithin legal discourse.22 I shall, therefore, work from (overly) simpleaxioms, by treating the answer to the following two questions as settled.First (starting from Samuel’s object of scrutiny): Is Gaius for the law whatDarwin was for science? Can, in other words, law be likened to zoology orany other natural science? I simply assume that the answer is: No,obviously not. Of course, the philosophy of science has brought us theinsight that neither scientific taxonomies23 nor even maps24 are justrepresentations of the real world, but always a construction as well.Scientific taxonomies are not different from legal ones simply because the

19 R Pound, ‘Classification of Law’ (1924) 37 Harvard Law Review 933–69.20 Waddams, above n 3, at 222ff.21 G Samuel, ‘Can Gaius Really Be Compared to Darwin?’ (2000) 49 ICLQ 297. See also

Samuel, above n 7, and Samuel above n 3.22 For the same reason, I shall not engage in an analysis of the methodological steps

(conceptualisation, categorisation, classification) involved in building a taxonomical ‘system’and shall not differentiate between terms such as ‘categorisation’ and ‘classification’ with thetechnical denotations assigned by other disciplines. See, for such definitions, R Ellen,‘Introduction: Categories, Classification and Cognitive Anthropology’ in R Ellen (ed), TheCategorical Impulse, Essays in the Anthropology of Classifying Behaviour (New York,Berghahn, 2006) 1.

23 This idea is usually illustrated by referring to Michel Foucault’s comment (Les mots etles choses: Une archeologie des sciences humaines (Paris, Gallimard, 1966) 7) on Borges’sapocryphal ‘certain Chinese encyclopedia’: ‘Dans l’émerveillement de cette taxinomie, cequ’on rejoint d’un bond, ce qui, à la faveur de l’apologue, nous est indiqué comme la charmeexotique d’une autre pensée, c’est la limite de la nôtre: l’impossibilité nue de penser cela.’ Foran English translation, see M Foucault, The Order of Things (New York, Random House,1973) xv. The understanding of taxonomies as relative to culture rather than correct orincorrect depictions of reality makes them a favourite subject of anthropologists and linguists;see, eg, Ellen, ibid and the study by G Lakoff, Women, Fire, and Dangerous Things, WhatCategories Reveal about the Mind (Chicago, University of Chicago Press, 1990). On the otherhand, cognitive psychology has insisted that categorisation of ‘the real world’ does nothappen completely arbitrarily; see, eg, CB Mervis and E Rosch, ‘Categorisation of NaturalObjects’ (1981) 32 Annual Review of Psychology 89, 91ff.

24 See, eg, JB Harley, ‘Deconstructing the Map’ (1989) 26 Cartographica 1–20.

28 Helge Dedek



former depict things that physically exist, whereas the latter order abstractconcepts that only exist in our minds. However, the categorisation of legalrules is special due to its normative implications. Categorising a rule, forexample, as ‘contractual’ rather than ‘tortious’ invokes a plethora ofdenotations25 that characterise, consciously and subconsciously, the cat-egory of ‘contract’: exercise of free will and consent as opposed toinvoluntary injury, distributive rather than corrective justice and so on.These properties of the category, in turn, inevitably influence the way aparticular rule—once placed in this category—is understood, construedand justified. Categorisation and classification in the law—and this is athought we will return to later—is not just descriptive, but prescriptive aswell. We cannot treat legal concepts as entities that can be ‘objectively’classified by family resemblance, needing simply to be spiked on the properpin to be displayed in a cabinet.

Given this answer, is it useless or even, as some scholars seem to think,ridiculous to work with the idea of a taxonomy, a classification of legalconcepts?26 Again, the answer is: No, obviously not. Categorisation andclassification are not restricted to ‘taxonomic’ disciplines such as zoology.Rather, they are the basic cognitive processes that permit the brain toreduce the complexity of stimuli to a degree that it can make sense of theworld.27 These processes are the precondition of the apprehension of thenatural world as well as the world of our own thoughts. Withoutcategorisation and classification, there would be no language to talk aboutlaw in the first place; without some kind of taxonomy, lawyers could notcommunicate, nor could laws be compared. Most importantly, taxonomiesare indispensable to legal education; let us not forget that Gaius’s as well asBlackstone’s mapping projects were driven by their pedagogical ambitions:‘A plan28 of this nature,’ writes Blackstone, ‘if executed with care andability, cannot fail of administering a most useful and rational entertain-ment to students of all ranks and professions.’29 Once we understand andacknowledge the instrumental rather than ‘natural’ character of legalcategories and classifications, there is nothing wrong with developingcategories and classifications that soundly serve their purpose. RoscoePound summarised it thus:

25 The idea that a multitude of information—defined by ‘communal traditions andcommunal processes’—finds expression in certain ‘paradigms’ that guide legal classification isalso expressed by JM Feinman, ‘The Jurisprudence of Classification’ (1989) 41 Stanford LawReview 661, 699.

26 D Campbell, ‘Classification and the Crisis of the Common Law,’ book review of TheClassification of Obligations by P Birks (1999) 26 Journal of Law & Society 369, 370:‘perfectly laughable’.

27 See Lakoff, above n 23, at 5ff; Ellen, above n 22, at 1ff.28 Emphasis in the original.29 W Blackstone, Commentaries on the Laws of England, facsimile of 1st edn 1765–1769

(Chicago, University of Chicago Press, 1979) vol 1, 36.




Classification is a shaping and developing of traditional systematic conceptionsand traditional systematic categories in order to organize the body of legalprecepts so that they may be: (1) Stated effectively with a minimum of repetition,overlapping, and potential conflict, (2) administered effectively, (3) taughteffectively, and (4) developed effectively for new situations.30

B. Demarcating Contract Law’s Territory

The title of this volume is ‘Exploring Contract Law’; given the historicalaffinity of ‘exploration’ and ‘mapping,’ it is not surprising that also in thisvolume we should find a contribution that engages in the enterprise of‘mapping’ contract law.31 The activities of ‘exploring’ and ‘mapping’ arenot limited to terra incognita. Indeed, Stephen Smith makes a compellingargument that some areas that are commonly considered to belong to‘contract law’ are not specifically ‘contractual’ at all. Rather, writes Smith,perhaps certain domains of contract law could be seen as belonging to ageneral part of the law of obligations. While we thought we knew themap—maybe not, to speak with Blackstone, down to every inconsiderablehamlet32 (this very fact, however, is what keeps the cartographers going!)—Smith redefines the whole territory, redraws the borders and lines ofdemarcation, and proclaims independence for an area hitherto unrecog-nised by the common law.

As I said, I do not want to engage with the question of whether thisenterprise is a legitimate or a useful one. Let us just assume that it is, andlet for now a simple justification suffice, namely, that thinking about acertain doctrine in a different way—for example, changing itsclassification—may shed new light on the purposes it serves. At the veryleast, it might help us to question the traditional explanations andjustifications we are taught to believe in. This critical inquiry, therefore, isa process that has an intellectual value in itself.33 In this article, however, Iwant to point to some of the problems that a classification, in particular adivision into the ‘general’ and the ‘special’ might entail. I would like todemonstrate how a classification that excludes certain areas from contractlaw as such runs the risk of obscuring certain connections that can only beseen when looking at contract law in a more inclusive way.

I want to do so by assuming a comparative perspective and reporting onthe experience of German civil law. Not being familiar with a comprehen-sive category of contract law, German civil law has, as a consequence,

30 Pound, above n 19, at 944 (emphasis in original).31 SA Smith, ‘The Limits of Contract’ ch 1.32 Blackstone, above n 29, at 35.33 The procedural character, the ‘process nature’ of classification is also mentioned by

Feinman, above n 25, at 704.

30 Helge Dedek



encountered certain doctrinal difficulties, particularly in justifying theaward of what is usually called ‘reliance damages,’ which a less fractionedview could have helped to avoid. This thought will lead us to the morefundamental problems associated with taxonomising the law, namely, thedanger of the proverbial ‘hardening of categories,’ that ‘well-knownailment of lawyers’.34

II. DIFFERENT TRADITIONS, DIFFERENT MAPS: THE ‘GENERAL’ ANDTHE ‘SPECIAL’

The move for a taxonomic ordering of the common law has at all timesbeen to some extent inspired by civilian ideas. As Smith acknowledges, thecartography he suggests displays, just like the Birksian writings on tax-onomy, a certain affinity to civilian ideas.35 In some sense, Smith goes evenfurther: his proposition is beyond just civilian—it is almost German. Theidea that contract law should not only be subdivided but also curtailed in away that makes room for a ‘general part of obligations’ and even a ‘generalpart’ of private law as such is a particularly German–civilian categorisa-tion. The taxonomists of the common law have also long been familiarwith the technique of classifying into ‘general’ and ‘special,’ the mostprominent of these taxonomists being, of course, Jeremy Bentham.36

However, many common law writers who have tried to make use of thistheme to guide their taxonomic efforts have explicitly looked at theclassificatory structure of the German civil law.37 This structure deviatesfrom the classical Gaian institutional scheme and from the organisation ofthe Code Napoléon as well.

A. Civil Law v Civil Law: The Hallmarks of German ‘Legal Science’

By ‘not only civilian, but German’ I also mean that certain hallmarks ofcivilianism, in particular the tendency to formalism and taxonomisation,

34 Lord Nicholls in Blake, above n 6, at 264 (HL) 284, quoting JP Dawson, ‘Restitutionor Damages’ (1959) 20 Ohio State Law Journal 175, 187.

35 Smith, above n 31 at fn 1.36 J Bentham, ‘Papers Relative to Codification and Public Instruction: Including Corre-

spondence with the Russian Emperor, and Divers Constituted Authorities in the AmericanUnited States’ in P Schofield and J Harris (eds), ‘Legislator of the World’: Writings onCodification, Law, and Education (Oxford, Clarendon Press, 1998) 8.

37 See, eg, A Kocourek, ‘Classification of Law’ (1934) 11 New York University LawQuarterly Review 319, 327: ‘The German Code has become the model of classification of ourtime.’ After sketching a taxonomy of taxonomies, Kocourek attempts to develop his own‘scientific classification,’ involving an elaboration of a ‘Theory of a General Part,’ comparinghis ideas to the actual organisation of the German code: ibid, at 341ff.




have been pushed to an extreme in the German civil law tradition, muchfurther than in France. The differences among the various expressions ofthe civilian tradition must not be underestimated. This is, of course, a tritestatement, but given how many stereotypes circulate about ‘the’ civil law, itis sometimes helpful to remember that the respective national civiliansystems are not alike in the way that common law systems are alike in theCommonwealth. Civilian systems may share the common past of theRoman law-based ius commune and, at least on the Continent, theexperience of codification. However, we must not forget that codificationitself ingrained a positivistic solipsism which had the consequence ofalmost completely cutting off discursive exchange between national legalsystems on the Continent. Furthermore, the respective codes are alreadythe product of different philosophies, styles, fashions and of the growingimportance of the process of ‘nationalising’ the law: the ‘age of codifica-tion’ was the same age that, on the Continent, witnessed the birth of themodern nation-state.38

If we compare the French and the German codes and try to describe theirdistinctive features, we are immediately confronted again with the perils ofclassification, this time in a different context: history (and the socialsciences in general) has to work with conceptual classifications such as‘epochs’ and ‘ideas’. Not unlike what the jurist tries to do by conceptuallycontextualising doctrines, rules and principles, the historian tries to explainpast events by fitting them into a broader context and by grouping themtogether under certain labels.39 The result is, again, ambivalent. Epochsand periods might indeed have something like a certain spirit that was in away characteristic and unique, and pointing out this specific property helpsone better understand historical events. However, by branding a certain erawith a certain mark we also construct an oversimplified identity anddifference that does injustice to continuities and similarities.

Keeping this in mind, we might say that the French Code Napoléon grewout of the eighteenth century spirit of Natural Law, expounding broadprinciples and making fundamental statements about justice and citizens’rights; whereas the German Code, a century younger, breathes the techno-cratic formalism of the nineteenth century, an age that endowed the idea of‘legal science’ (with all its implications for legal taxonomy and classification)

38 On the nationalisation of the ius commune (the basis of the codes!) by nationalisinginstitutional writing see K Luig, ‘The Institutes of National Law in the Seventeenth andEighteenth Centuries’ (1972) 17 Juridical Review (NS) 193, 195: ‘This new law was acomplete and coherent unit, and theoretically independent of the earlier European iuscommune; it was the expression of the national independence and autonomy of the risingnation-states of the ancien régime.’

39 See, eg, C Behan McCullagh, ‘Colligation and Classification in History’ (1978) 17History and Theory 267ff.

32 Helge Dedek



with a whole new meaning.40 Of course, this would be an over-simplification.The ‘scientific’ approach to law and the development of taxonomic schemeswas not an invention of the nineteenth century; for the longstanding nature ofthis ‘scientific’ tradition, we might point to Leipniz, the mos geometricuschampioned by the enlightenment in general,41 to Petrus Ramus,42 or wemight go back to the tables of distinctions of the medieval scholastics43 andfinally, of course, to Gaius himself. However, the approach to law did changeagain in an age that saw industrialisation and the rise of the market andbelieved in progress through the advancement of science and technology.Savigny’s Historical School, which proclaimed that the ius commune could beimproved by an informed study of the classical Roman sources, employingstate-of-the-art methodologies of linguistics and history, turned slowly into‘the science of the Pandects’44 whose ideal of the law was that of animmaculate clockwork. Cutting-edge scholarship had, if you will, given wayto engineering. High-end engineering, that is: centuries of law as an academicdiscipline and an ideal of intellectual rigour befitting the professor (just recallMax Weber’s ‘Science as a Vocation’) drove German nineteenth century legalscholars to push their conceptual jurisprudence to an unparalleled degree ofmechanical precision and technical perfection. While again walking the thinline between attaching a helpful label and falling victim to cliché, we mightsay that this expresses, better than anything else, the ‘Germanness’ of Germanprivate law—a veritable field day for those who see connections between legalculture and national character.45 However, if you are interested in how ajurisprudential style that is safely grounded in a highly developed ‘scientific’taxonomy operates, looking at German law is definitely the right thing to do.

40 See, eg, K Zweigert and H Kötz, An Introduction to Comparative Law, 3rd edn, transT Weir (Oxford, Clarendon Press, 1998) 145.

41 See, eg, R Berkowitz, The Gift of Science (Cambridge, Harvard University Press, 2005)17ff.

42 JS Freedman, ‘The Diffusion of the Writings of Petrus Ramus in Central Europe, c1570–c 1630’ (1993) 46 Renaissance Quarterly 98ff. See also DR Kelley, ‘JurisconsultusPerfectus: The Lawyer as Renaissance Man’ (1988) 51 Journal of the Warburg and CourtauldInstitutes 84, 91ff.

43 G Otte, Dialektik und Jurisprudenz, Untersuchungen zur Methode der Glossatoren(Frankfurt am Main, Vittorio Klostermann, 1971) passim.

44 F Wieacker, Privatrechtsgeschichte der Neuzeit, 2nd edn (Göttingen, Vandenhoeck &Ruprecht, 1967) 430.

45 See, eg, OG Chase, ‘Legal Processes and National Culture’ (1997) 5 Cardozo Journalof International & Comparative Law 1 (and see of course JH Langbein’s angry reply‘Cultural Chauvinism in Comparative Law’ (1997) 5 Cardozo Journal of International &Comparative Law 41). For a far more subtle elaboration on the ‘Frenchness’ of French civillaw see Legrand, above n 16, at 235ff; N Kasirer, ‘Larger than Life’ (1995) 10 CanadianJournal of Law and Society 185. National peculiarities were of course tempered by actualintellectual exchange: the ‘Pandectistic’ style of jurisprudence was, in the nineteenth century,of great allure for all civilians (and, incidentally, for many common lawyers as well!) and ofmajor influence in jurisdictions that had adopted ‘Natural Law Codifications’ earlier, such asPrussia and Austria, and even France: French civil law scholarship was ‘pandectised’. Let usnot forget that the canonical Aubry/Rau started out as a Pandectist treatise.




Whether one sees a pinnacle of plain orderly elegance or a pedantic nightmareof anachronistic formalism is in the eye of the beholder alone.

B. Stephen Smith’s ‘Limits’ vis-à-vis the Limited German Concept of‘Contract Law’

(i) Horizontal and Vertical Borders

Having refreshed this bit of background information, we can turn back tocontract law. In his contribution to this volume, Stephen Smith distin-guishes two sets of borders that demarcate contract law’s territory. First,there is a ‘horizontal’ border, which ‘separates contract law rules from theother sets of primary obligation-specific rules that occupy the bottom ofthe obligations pyramid’46 as, for example, the obligation not to trespass,the obligation not to damage another’s property, the obligation not todefame and so on. Secondly, Smith defines the ‘vertical’ border of contractlaw as the one that

sets the boundary between contract law and the general part of the law ofobligations. This border is vertical because it sets the boundary between rules atthe top and bottom of the pyramid; specifically it distinguishes contract law rules(and other primary-obligation rules) from rules that are located above them inthe sense that they qualify these rules or regulate the consequences of breachingthe obligations to which these rules give rise.47

To find out whether a rule ‘vertically’ belongs to the law of contracts,Smith asks two questions: ‘(1) is the rule applied to resolve not justcontractual but also non-contractual disputes? (2) is the rule of a kind thatit makes sense, morally, to suppose should be applied to obligationsgenerally as opposed merely to contractual obligations?’48 Asking the latterquestion implies an assumption we briefly mentioned before, namely, thattaxonomy in law is never only descriptive, but also prescriptive in nature.

(ii) The ‘Law of Obligations’ in the German Civil Code

Smith acknowledges explicitly that his call for ‘vertical borders’ bears acertain resemblance with civilian ideas and, in particular, the German wayof defining the scope of contract law in respect to a ‘general part’ of the‘law of obligations’.49 Indeed, the German Civil Code is built around theprinciples that repetition should be avoided, and that rules that are to be

46 Smith, above n 31, in the text accompanying fn 4.47 Ibid, text accompanying n 5.48 Ibid, in the text at end of Section III.49 Ibid, in the text accompanying n 1.

34 Helge Dedek



applied generally and not just in one particular context are to be system-atically grouped together to form a ‘general part’.50 Thus, the codalchapter (‘book’) on the law of obligations is divided into subsections,whose ambit moves from the rules applicable to all obligations, to thoserules applicable to all contractual obligations, and then, finally, to the rulesonly applicable to ‘specific obligations.’ Only the eighth subsection dealswith specific obligations: it is subdivided into 27 ‘titles’ which first dealwith specific contracts (sale, loan, donation and so on) and then, in the lasttwo titles, with unjustified enrichment and delicts. Rules that are supposedto be even more general in the sense that they are to be applied not only inthe law of obligations but also to all other parts of the Code (the ‘law ofthings’, family law, hereditary law) are clustered together in the ‘GeneralPart,’ §§ 1–240, as are, for example, the rules on personhood, age ofconsent and so on.

(iii) Breaking Away from Gaius: Origins of the ‘Pandektensystem’

This regulatory style, moving from the general to the specific, culminatingin the overarching abstractness of the ‘General Part’, has been seen as theunique characteristic of the German civil law.51 The underlying techniquehas been likened to the mathematical operation of ‘factoring out’, isolatinga factor in an equation and placing it in front of parentheses.52 Asexpounded by Andreas Schwarz in the still canonical treatment of thesubject,53 the historical background of this ‘Pandektensystem’ and itsparticular ‘mathematical’ approach is the osmosis of Natural Law ideasregarding the systematisation of law into the scholarly treatment of the iuscommune. Natural lawyers Pufendorf and then, most notably, ChristianWolf, developed ‘logical’ legal systems in the abstract, which, althoughwithout any significance for the law in action, inspired ius commune

50 For a concise introduction to the ‘regulatory technique’ of the BGB see BS Markesinis,The German Law of Contract, 2nd edn (Oxford, Hart Publishing, 2006) 16ff. For anintroduction to the historical reasons for this particular approach, see R Zimmermann, TheLaw of Obligations (Oxford, Oxford University Press, 1996) 29ff. See also Kocourek, aboven 37, at 327.

51 See, eg, F Pollock, Essays in Jurisprudence and Ethics (London, Macmillan, 1882) 14(looking at Pandectist writing). This technique has, taking part in the process of ‘Pandectisa-tion’ on the Continent (see above n 45), influenced other civilian traditions as well, even if thegeneral–special dichotomy is not reflected in their respective codes. Pound, therefore, calls thearrangement that singles out a ‘general’ part that of ‘the modern Civilian’ (Pound, above n 19,at 947, 962ff).

52 This common metaphor seems to have been first employed by G Boehmer, Grundlagender Bürgerlichen Rechtsordnung, (Tübingen, JCB Mohr (Paul Siebeck), 1951) vol 2, pt 1, 72.

53 AB Schwarz, ‘Zur Entstehung des modernen Pandektensystems’ (1921) 42 Zeitschriftder Savigny-Stiftung für Rechtsgeschichte Romanistische Abteilung 578.




scholars to deviate from the traditional institutional scheme.54

Introduced—but later abandoned again—by Gustav Hugo in his treatise‘Institutionen des heutigen römischen Rechts’ (1789), it was fully devel-oped by Georg Arnold Heise in his work ‘Grundriß eines Systems desgemeinen Civilrechts zum Behuf von Pandecten-Vorlesungen’ (1807).However, we do not know whether this ‘Pandektensystem’ would havebeen immortalised in the structure of the Civil Code a century later, had itnot been for the endorsement of the most prominent heads of Germanprivate law scholarship,55 among them Friedrich Carl von Savigny himself,the towering figure of German nineteenth century legal scholarship.Savigny adopted Heise’s system for his lectures and thus helped to make itthe standard order in which the structure of the German ius commune wasunderstood.56 This perception then determined the structure of the Code;to this very day, it provides the outline that defines the ‘classical’ scope oftreatises on private law and, most importantly, the curriculum of privatelaw education.57 General Part, General Part of the Law of Obligations,Specific Part of the Law of Obligations (divided into contractual/non-contractual obligations) is still the typical cursus law students have to passthrough. The taxonomy, thus, moulds and dominates the mindset of eachand every lawyer.

(iv) The Interplay of the General and the Special: Examples andProblems

Let us try to understand the effects of this compartmentalised approach bylooking, for reasons of convenience, at some of the examples treated bySmith. We already briefly addressed the rules on personality. These rulesapply not only in the context of contractual obligations, but determine,according to the logic of the Bürgerliches Gesetzbuch (BGB), whether aperson—legal or ‘natural’—can be the subject of any kind of right, be it inthe context of contract, property or family law. Therefore, according to the

54 Ibid, at 583. French law, of course, has seen similar developments, but with differentconsequences for the code. See, most notably, Domat’s Les droits civiles dans leur ordrenaturel (1689–97).

55 Schwarz, above n 53, at 580.56 M Avenarius, ‘Der Allgemeine Teil des Obligationenrechts aus Savignys Pandekten-

manuskript-Bedeutung und Grundsatze der Edition’ in M Avenarius (ed), Friedrich Carl vSavigny, Pandekten, Obligationenrecht, Allgemeiner Teil, (Frankfurt am Main, VittorioKlostermann, 2008) 12ff.

57 It is interesting that Pound (above n 19, at 940) remarked disapprovingly on the factthat, vice versa, the organisation of the Code had been influenced by the pedagogicallymotivated arrangement of Windscheid’s famous textbook on the Pandects, which, accordingto Pound, might not have been the optimal arrangement for a restatement of the law.However, properly understood, Windscheid’s treatise was nothing less than a restatement ofthe law in itself.

36 Helge Dedek



taxonomy of the BGB, these rules are located not in the general part of thelaw of obligations, but in the ‘General Part’ of the whole Code.

(a) Formation of Contract There, in the ‘General Part,’ we also find therules on mistake, duress, unconscionability and illegality. This issurprising—why would those doctrines be treated outside the law ofobligations? Their placement is owed to the fact that the rules on ‘legaltransactions’ (Rechtsgeschäft) and ‘declarations of will’ (Willenserklärung)are located here, which actually means that the rules on offer andacceptance and the ‘meeting of the minds’ are not dealt with as part of thelaw of obligations at all. This, of course, is even more surprising—are notthose rules, of all rules, the most ‘contractual’ ones? Yes, they are. But inaspiring to distill the general ideas out of specific rules and principles,German private law developed a general concept of ‘contract’ beyond justcontracts that create obligations. This, of course, must seem utterlyenigmatic to the common lawyer who is not even able to see a ‘gift’ as a(donative) contract because it does not entail promises of future perform-ance. The most prominent example of the non-obligatory contract wouldbe an agreement that transfers ownership.58 In German law, an obligatorycontract, such as a contract of sale, does not per se transfer ownership. Thetransfer of ownership requires the transfer of (corporeal) possession(traditio or an equivalent), and, in addition, another separate agreement inwhich the parties agree on the transfer of ownership. This second agree-ment does not create an obligation, but alters legal status—in our example,the allocation of ownership. This agreement would still be seen as a‘contract.’ With contract thus defined, it makes sense (if only according tothe twisted logic of the unique understanding of ‘contract’ in German law)to place the rules on formation of contract in the parts of the Code that aresupposed to be applicable not just to contract law, not just to obligations,but to all agreements in private law, be they the source of obligations ornot.

In this specific institutional context, the first part of Smith’s twofold testof whether a rule should be seen as ‘general’ or not would be satisfied.What about the second part of Smith’s test: does it make good moral senseto treat it as a general rule? This is obviously a more complicated question.If we had the chance to ask the historical spiritus rectores of this

58 See, on the so-called ‘Principle of Separation’ and the ‘Principle of Abstraction’ thatdefine the ‘relationship of obligation’ and the ‘transaction that transfers, alters, [etc], rights’ asseparated and independent of each other, Markesinis, above n 50, at 27. The broad concept of‘contract’ beyond agreements giving rise to obligations was developed by Friedrich Carl vonSavigny, System des heutigen römischen Rechts, Bd III (Berlin, Veit und Comp, 1840) § 140,312ff. On the significance of this development for the placement of ‘legal transactions’ in the‘General Part’ see JF Stagl, ‘Die Rezeption der Lehre vom Rechtsgeschäft in Österreich durchJoseph Unger’ [2007] Zeitschrift für Europäisches Privatrecht 37, 39.




taxonomy, some might be confused by the implication that the functioningof the clockwork would have anything to do with morality.59 It is in thissense that Smith transcends his predecessors in the taxonomic enterprise,who would have described their project as exclusively involving logics, notmorals.

However, others, and among them the titanic Savigny himself, wouldhave answered, I am sure, in the affirmative as follows: the capacity of aperson to bring obligations into life and to create, transfer and dispose ofrights by their sheer willpower is more than contract law mechanics, it isan expression of the autonomy of the will as such. The autonomous will isa condensation of the idea of individual freedom, which is the centralmoral concept in early nineteenth century German private law.60 Privatelaw demarcates spheres of individual freedom in a way that ‘assigns theindividual will a region where it can reign unperturbed by any other will’.61

According to this ideology, the technical rules which are the most immedi-ate emanation of this principle deserve to be exposed in the ‘General Part’as a statement about the fundamental significance of the will in privatelaw. Having crystallised in the structure of the Code, this ideology is alsoset in stone and remains so in a changing intellectual environment.62 Thequestion suggested by Smith has to be asked again and again to prevent the‘hardening of categories’ into anachronistic fossils.

(b) Contract Damages The codal classification of the rules on damagesdepends, likewise, on the degree of their generality. Being applicable onlyto obligations, sedes materiae of those rules is not the ‘General Part’.However, rules that govern contractual as well as tortious obligations, likethe principle of the restitutio in integrum, are deemed to be part of the‘General Part of the Law of Obligations’. Rules that expound the peculi-arities of remedies typical of certain specific obligations—such as thereduction of the purchase price as a remedy available to the buyer or the

59 On those writers who adopted the system for no other reasons than that it was awell-entrenched tradition see Schwarz, above n 53, at 580. On the ‘secular agnosticism’ of thenineteenth century legal scientists and their refusal to make moral commitments see, eg, JGordley, The Philosophical Origins of Modern Contract Doctrine (Oxford, Clarendon Press,1991) 227.

60 On ‘will theory’ thus understood as synecdoche for the nineteenth century ‘legalconsciousness’ see D Kennedy, ‘Two Globalizations of Law & Legal Thought, 1850–1968’(2003) 36 Suffolk University Law Review 631, 637. On how far the idea that this ‘legalconsciousness’ is rooted in Kantian or Hegelian philosophy is still a matter of contention, see,for an introduction, H Dedek, Negative Haftung aus Vertrag (Tübingen, JCB Mohr (PaulSiebeck), 2007) 101. See also Gordley, ibid, at 227.

61 Savigny, above n 58, at vol 1, § 52, 333: ‘[…] dass dem individuellen Willen ein Gebietangewiesen wird, in welchem er unabhängig von jedem fremden Willen zu herrschen hat.’

62 For an example of the outrage that was provoked when provisions on consumerprotection were implanted in the General Part, see W Flume, ‘Vom Beruf unserer Zeit fürGesetzgebung’ [2000] Zeitschrift für Wirtschaftsrecht 1427.

38 Helge Dedek



payment of damages in periodic installments in case of a delictual injurythat led to bodily harm—would be found in the context of the respectivenominate contracts or the sections on delicts.

This scatteredness of the rules on damages draws our attention toanother problem the undertaking of inducting general principles fromparticular rules faces. Abstract general principles do not apply in theabstract. General principles apply to particular cases. Damages areawarded if the breach of an obligation causes loss. However, the obligationin question is never just the pure, abstract and universal concept of anobligation stripped of all its worldly accidentia,63 but is a contractualobligation or a tortious obligation or a restitutional obligation, and so on.If we are trying to distill general principles from all these different fields,we are running the risk that the common denominator is so insignificant,the general principle so abstract that it is actually too abstract to be appliedin real life without adjustments to the respective context. The consequenceis that, if a set of rules is classified as ‘general’, there is a need foradditional rules that assure the adjustments of the general rules to thespecific contexts of sales, torts, unjustified enrichment and so on.

Smith encounters the same difficulty when reflecting on a generalprinciple of responsibility: surely, there is a general idea that underlies ruleson the vitiation of contractual consensus as well as the rules on theresponsibilities for one’s wrongdoing. The concept of responsibility may begeneral, but ‘duress and other general defences do not operate identicallyacross contractual and non-contractual settings’.64 There obviously is atheme of responsibility pertaining to both the law of contractual andnon-contractual obligations, but its actual emanations in both fields are sodifferent that it seems impossible to formulate a general rule—henceSmith’s scepticism as to the classifiability of this principle of responsibilityas belonging to the ‘General Part of the Law of Obligations’.

The German Civil Code, keener on creating ‘General Parts’ whereverpossible, has to deal with exactly this effect of creating repetitions insteadof avoiding them.65 For every general principle there have to be many otherrules that adapt the general principle to special circumstances. The result is

63 At least in regard to applying the rule to an actual factual situation; however, this is notnecessarily meant to imply that, in the Nominalist sense, such a concept could not exist at all.

64 Smith, above n 31 in the text accompanying fns 46 and 47.65 This malaise becomes particularly apparent when analysing the German Civil Code,

but is, of course, a problem other codifications have to put up with as well. See, eg, the ruleson ‘lesion’ in the Quebec Civil Code: Art 1406 ss 1 CCQ 1 defines ‘lesion’ as a principle, ss 2adjusts the principle for cases involving minors or ‘protected persons of full age.’ Art 1405,however, explains that ‘lesion’ vitiates consent exclusively in respect to minors, to saidprotected persons, and the cases expressly provided by law. These cases are limited toextraordinary situations as, eg, addressed in Art 424. Against this backdrop, expounding the‘general principle’ seems of rather limited value and can only be explained by its historicalroots. See P-G Jobin, Les Obligations, 6th edn (Cowansville, Yvon Blais, 2005) 267ff.




that provisions dealing with the same subject matter are scattered all overthe Code, as we just saw in the case of the rules on damages. This makes itvery difficult to get a coherent picture of the law: the stones of the mosaichave to be pieced together somewhat tediously to be able to see a coherentpicture. Or, if we want to return to the imagery of the clockwork: theGerman Civil Code turns out to be a veritable ‘Grande complication’whose mechanism only a most accomplished master clockmaker couldunderstand in its full complexity.66

The recent reform of the German law of obligations67 has even furtherincreased this effect. In an attempt to erase anachronistic peculiarities likethe ‘Aedilitian’ remedies available to the buyer in sales contracts (remnantsof the Roman law of market transactions, administered by the magistratesin charge of market exchanges, the aediles, not the praetor68), remediesacross all nominate contracts were supposed to be harmonised and,therefore, the importance of the ‘General Part of the Law of Obligations’heightened. The technical means employed was a web of complex crossreferences cast over the Code, referring to the ‘General Part of the Law ofObligations;’69 in turn, another set of rules had to be introduced to adjustthe general rules to the context of the different types of contracts.

Take, for example, a buyer’s claim for damages ‘in lieu of performance’in case of the sale of a defective good. Let us assume that a contract hasbeen concluded according to the rules on offer and acceptance in § 116BGB and the following paragraphs, that this contract is a contract of salesas defined by § 433 BGB, and that the good delivered is defectiveaccording to the definitions of § 434 BGB.70 The buyer’s claim fordamages, her ‘subjective right’, would be seen as rooted in § 281 BGB, a‘general’ rule on damages for breach of an obligation, which is entitled‘Damages in lieu of performance for non-performance or failure to render

66 See also Zimmermann, above n 50, at 31.67 For a comprehensive introduction, see R Zimmermann, The New German Law of

Obligations (Oxford, Oxford University Press, 2005).68 See Zimmermann, above n 50, at 311ff. For the reconstruction of the edict see O Lenel,

Das Edictum perpetuum: Ein Versuch zu seiner Wiederherstellung, 3rd edn (Leipzig,Tauchnitz, 1927) 554ff; É Jakab, Praedicere und cavere beim Marktkauf—Sachmängel imgriechischen und römischen Recht (München, CH Beck, 1997) 123ff, 153ff, 272, 291ff, 296.

69 See, eg, B Dauner-Lieb, ‘Die geplante Schuldrechtsmodernisierung’ [2001] Juristenzei-tung 8, 12ff, 16; H Dedek, ‘Vorbemerkung zu §§ 241ff’ in M Henssler and F Graf vWestphalen (eds), Praxis der Schuldrechtsreform, 2nd edn (Recklinghausen, ZAP-Verlag,2002), passim.

70 § 434 BGB partly incorporates the standards set by the Directive 1999/44/EC of theEuropean Parliament and of the Council of 25 May 1999 on certain aspects of the sale ofconsumer goods and associated guarantees (OJ L 171 of 7 July 1999 at 12), generalising thedefinitions meant for the sale of consumer goods for all sales contracts.

40 Helge Dedek



performance as owed’,71 and whose purpose is to decide whether thecreditor of the breached obligation has access to damages on the basis ofan expectation measure, replacing her initial right to specific performance.§ 437 Nr. 3 BGB, entitled ‘Rights of buyer for defect,’ forms the jointbetween ‘general’ and ‘special’ parts and declares that § 281 BGB isapplicable as well in cases of sale of defective goods. § 437 Nr. 3 BGBexplicitly refers to § 281 BGB, which is, in turn, a modification of theprinciple stated in § 280 BGB, which announces that damages have to bepaid for breach of obligations, but only if the breaching party was at fault.To find out whether this is the case, we now need to look at the scope ofobligations of the seller under a sales contract, which is again defined inthe ‘special part’, in § 433 BGB. If we find that indeed an obligation hasbeen breached, we jump back to § 276 BGB to consider the standards thatdecide whether the seller was at fault. If the requirements of § 280 BGB arethus satisfied, we return to § 281 BGB in order to check whether its nextimportant requirement has been met: in order to replace the right tospecific performance with a claim for damages, the creditor needs to givethe debtor a warning and a second chance to perform properly within areasonable timeframe. In certain cases, enumerated in § 281 section 2BGB, such a warning and notice is dispensable, for example, in cases offraudulent misrepresentation72 or repudiation. However, if none of theseexceptions apply, we have to—as if following a tennis match—shift ourgaze back to the special rules on sales contracts again. There, we find §440 BGB, which explicitly extends the list of exceptions expounded in §281 section 2 BGB in order to adjust it to the specific interests of theparties to a sales contract. If we, finally, come to the conclusion that thebuyer is indeed entitled to damages, the quantum would be calculatedaccording to the guidelines found in the ‘General Part of the Law ofObligations’ again: § 281 BGB, ordering the expectation measure ofdamages ‘in lieu of performance,’ and § 249 BGB and following para-graphs, containing the rules around the general principle of restitutio inintegrum.

Confusing as it might seem, the way I described the back and forthbetween the rules in the different parts of the Code is a gross oversimplifi-cation compared to the way a student—and eventually, a judge—would beexpected to actually treat the matter.73 Rather than sublime clockwork,

71 Translations of the now official titles of the respective provisions as offered on theofficial website of the German Federal Ministry of Justice: <http://www.gesetze-im-internet.de/englisch_bgb/englisch_bgb.html> (last accessed November 2008).

72 Bundesgerichtshof, Beschluss vom 8. 12. 2006 – V ZR 249/ 05 (2007) Neue JuristischeWochenschrift 835 citing Dedek, above n 69, at § 281, [36].

73 It would also be simple to present far more intricate and complex examples of thistechnique.




this mechanism seems more like a mental pinball machine that tosses youback and forth relentlessly between its special and its general parts.74

III. THE HARDENING OF CATEGORIES: THE CASE OF RELIANCEINTEREST IN CONTRACT DAMAGES

We have seen, so far, some problems that might arise in a system relyingtoo heavily on a division of law into categories of ‘special’ and ‘general’The ‘atomisation’ of law in such a system, however, is beyond just atechnical problem that makes the application of rules more tedious. The(‘vertical’) borders between the different ‘parts’ do not just place certainrules in certain ‘areas’, they create intellectual boundaries as well that, atleast over time, prevent jurists from seeing coherence and continuity aboveand beyond the borders. Seeing these particular connections can be asimportant as the ability to understand the generality of certain ideas. Whilecategorising, it is therefore vital to keep in mind the permanent possibilityof a different categorisation, and thus to keep open the recourse todifferent ways of legal reasoning. What we have to fear about categorisa-tion in terms of ‘general’ or ‘special’ is not so much the possibleincorrectness of such a categorisation, but its petrifaction—the ‘hardeningof categories’ that cuts off the access to valuable legal arguments.

Stephen Smith concludes that ‘failing to pay attention to contract law’svertical limits can lead scholars, judges, and lawyers to apply the wrongprinciples to understand legal rules, to draw the wrong inferences fromthose rules, and to fail to make appropriate generalisations’.75 One couldreply, from the perspective of a legal system built on rigorous classificatoryschemes, that this might go both ways: taking these limits too seriously canlead scholars, judges and lawyers to fail to draw appropriate connectionsbetween the concepts separated by the vertical borders. The important andcomplicated task would be to keep both perspectives, limited and border-less, at the same time.

Let me try to illustrate my point by taking a closer look at thephenomenon known as the ‘reliance interest in contract damages’. Thearticle so entitled by Lon Fuller and William Perdue76 is famed as beingprobably ‘the most influential single article in the entire history of modern

74 For a general criticism of the complexity of the German Civil Code, see Zweigert andKötz, above n 40, at 144ff. On the relationship between the ‘General Part’ and the pervasivefascination for abstract concepts, the hallmark of German ‘conceptual jurisprudence,’ see,most recently, Stagl, above n 58, at 39.

75 Smith, above n 31 in the first paragraph of Section VIII Conclusion.76 LL Fuller and WR Perdue Jr, ‘The Reliance Interest in Contract Damages: 1’ (1936) 46

Yale Law Journal 52.

42 Helge Dedek



contract scholarship, at any rate in the common law world’.77 Interestingly,the Fuller and Perdue article is being hailed particularly (among others) forits taxonomical achievements. For example, Grant Gilmore, in languageremarkably suitable to our exploratory theme, applauds Fuller for having‘discovered’ a ‘case law underground’ that turned out to provide theprotection of what he christened the ‘reliance interest’.78

According to Smith’s map, however, this famous article, by focusing ondamage rules, treats a subject matter that lies outside the realm of contractlaw.79 Even though a theory of the law of damages might, indirectly,provide helpful insights for the law of contracts as well,80 these territoriesare separated by a ‘vertical border’. This train of thought brings us,paradoxically, to another thrust of the article that was ‘taxonomical’ innature as well. Fuller and Perdue attack, from the outset, the understand-ing that, when we evaluate contract damages by applying an expectationmeasure, we are just applying a general rule guiding the award ofcompensatory damages, namely, the rule that the victim of an injury simplyis to be put in the same position, ‘as far as money can do it’, as if the injuryhad not occurred. The article calls this view into question; it tries to provethat the way we first formulate and then apply a ‘general rule’ ofcompensation to the situation of a breach of contract is already anexpression of what we think the purpose of contract law should be. In nowfamous words, the authors wrote:

Yet in this case we ‘compensate’ the plaintiff by giving him something he neverhad. This seems on the face of things a queer kind of ‘compensation.’ We can, tobe sure, make the term ‘compensation’ seem appropriate by saying that thedefendant’s breach ‘deprived’ the plaintiff of the expectancy. But this is in essenceonly a metaphorical statement of the effect of the legal rule. In actuality the losswhich the plaintiff suffers (deprivation of the expectancy) is not a datum ofnature but the reflection of a normative order.81

This highlights the structural interlock between awarding compensatorydamages according to the expectation measure and the purposes ofcontract law. We cannot just say there is institutionalised reliance in the

77 PS Atiyah, Essays on Contract (Oxford, Clarendon Press, 1986) 73.78 G Gilmore, The Death of Contract (Columbus, Ohio State University Press, 1974) 56.

See also TD Rakoff, ‘Fuller and Perdue’s The Reliance Interest as a Work of LegalScholarship’ [1991] Wisconsin Law Review 203, 206ff, who criticises Gilmore for stylisingFuller’s achievement as ‘discovery’, and instead points to the intellectual groundwork done byWilliston and Cohen.

79 Smith, above n 31, text accompanying fns 11–14: ‘Often regarded as the most famousarticle in English on contract law, Fuller and Purdue’s ‘The Reliance Interest in ContractDamages’ focuses, as the title states, on damages rules. Contract courses in common lawjurisdictions frequently begin with a discussion of damages. … But the rules governingcontract damages are not contract law rules.’

80 Smith, above n 31, in text accompanying fn 26.81 Fuller and Perdue, above n 76, at 53.




expectancy that justifies the treatment of an obligation as a property right,leading to the treatment of the loss of expectancy as an actual injury, whenon the other hand this reliance is grounded in the very fact that enforce-ment of contractual promises consists of awarding the expectancy. Intaxonomical terms, by showing this circular interdependence, ‘The Reli-ance Interest in Contract Damages’ tries to break down the conceptualboundaries between the ‘general law of damages’ and ‘contract law’; thearticle seeks to de-generalise our understanding of the ‘general’ rules thatpurport to be applicable to actual tortious injuries as well as to the ‘loss’suffered as a consequence of the breach of a contractual obligation. Itspurpose is to make us think about the award of damages in contract casesas lying at the very heart of contract law and to dispense with any verticalborders that might separate the law of contract damages from the realm ofcontract law.

Now, this might be, in terms of the doctrinal implications, a good idea ornot. This is not what is important for our inquiry. My point is that only ina system that is not founded on a deeply entrenched understanding of lawas fractioned into ‘general’ and ‘special parts’, is it at all possible to see thekind of coherence and interdependence that Fuller and Perdue do. Only insuch a system does the legal analyst have the freedom to transcendboundaries and recombine elements that one might see as separated byvertical borders; the freedom to go even one step further and proclaim that,as, for example, Patrick Atiyah did, the binding force of promise isgrounded in reliance alone and that, therefore, the enforceability ofpromises should express itself in the award of reliance damages as theprimary remedy.82

Again, the point is not whether this way of thinking has its doctrinalmerits or not.83 The point is that making such a connection would beliterally unthinkable within the partitioned framework of German civillaw.84 Under the German approach of strict ‘vertical’ separation, ‘contract’as a coherent institution is completely dissolved in the system. Formationof contract is a matter of the ‘General Part’. Once the contract is formed, itgives birth to obligations, which are dealt with by another part of themechanism, the ‘General Part of the Law of Obligations’. Here, theobligation is seen abstracted from its contractual origin and treated as anideal concept without an individual history. In case of a breach, the generalrules on damages apply (as we have seen, possibly modified by more

82 PS Atiyah, Promises, Morals, and Law (Oxford, Clarendon Press, 1981) 36ff; Atiyah,above n 77, at 20ff.

83 For a discussion of this question, see SA Smith, Contract Theory (Oxford, OxfordUniversity Press, 2004) 413ff; SA Smith, ‘“The Reliance Interest in Contract Damages” andthe Morality of Contract Law’, Issues in Legal Scholarship, Symposium: Fuller and Perdue(2001): Article 1, <http://www.bepress.com/ils/iss1/art1> (last accessed November 2008).

84 On the following, see Dedek, above n 60, at 6ff, 111ff.

44 Helge Dedek


JOBNAME: Neyers - Exploring C PAGE: 21 SESS: 5 OUTPUT: Fri Apr 17 10:40:29 2009

specific rules). The taxonomic scheme itself prescribes the application ofthe general rules on damages, thus equating loss of expectancy in contrac-tual and delictual constellations. The question of what amount of damagesis awarded becomes a mere mathematical operation based on the idea ofcausation, comparing the claimant’s actual financial position after theinjury to the hypothetical financial position he would be in but for theinjury. The amount by which the latter exceeds the former is deemed to bethe pecuniary loss ‘caused’ by the injury. Applying this formula, aninvestment made in reliance on a promise which turns out to be uselesswhen the promise is broken is, strictly speaking, not ‘caused’ by the breach:even if the contract had been performed, the investment would have beenmade and would have diminished the investor’s patrimony nonetheless. Interms of tangible, pecuniary loss, therefore, the hypothetical financialposition the claimant would be in but for the breach equals the actualposition after the breach took place. The mere fact that the investment hasbeen rendered futile has been seen as an intangible loss. Therefore, bymechanically applying this formula the ‘contractual’ measure of damagessimply does not entail reliance damages.

The very idea that reliance invested at the stage of the formation ofcontract could have any implications for the calculation of damages isinconceivable; the system simply has decided that both stages have nothingto do with each other. Once the system—the structure of the Code and thejurisprudence built on this structure—has petrified, it cannot escape itsself-imposed rigidity. It has to react by inventing new devices that add tothe complexity of the mechanism, for example, the reliance-based butextra-contractual institute of ‘culpa in contrahendo’.85

Courts, working with the formula that gains prevented by the breach arelosses that can be compensated, have employed an institutionalised factualassumption that investments made in reliance would have been cost-efficient—explicitly pointing out that the futile investment as such is notthe loss that has been incurred, but the loss of opportunity to make a profitthat would at least have equalled the amount of the investment.86 Becausethis differentiation was taken very seriously, new problems arose in caseswhere an actual intention to make a profit could not be shown.87 It washeld that in such cases the expectancy could not be measured by takingrecourse to out-of-pocket investments and, therefore, no damages could beclaimed at all.

Finally, the legislator reacted by implanting a new provision into theCode, § 284 BGB, granting a right to reliance damages. But even when the

85 See now §§ 311 ss 2, 280 BGB.86 Bundesgerichtshof BGHZ 71, 234, 238ff; BGHZ 123, 96, 99: the so-called ‘Rentabil-

itätsvermutung’.87 Bundesgerichtshof BGHZ 99, 182, 196ff: the so-called ‘City Hall Case’.




legislator himself sprang into action, German legal scholars claimed that aprovision granting reliance damages for breach of contract would be amonstrosity because it would violate the ‘general law of damages’ (as if itwere a law of nature), particularly the general rule that the aggrieved partycould only be compensated for losses that were caused by the breach,whereas this causal link was missing between the breach and the detrimen-tal change in position incurred in reliance on the promise.88

Obviously, the problems related to causation have occurred to commonlawyers as well: ‘The reliance interest fixes the wrong as making thepromise, not breaking it.’89 It is understood that by claiming reliancedamages, the aggrieved party is asking ‘not to be put into the position hewould have been in had the contract been performed, but to be put into theposition he would have been in had it never been made.’90 How is it, then,possible to award reliance damages for breach at all? It has been suggestedthat the award of reliance damages could be explained by the assumptionthat the disappointed party would have at least ‘broken even’91—whichdoes not, however, account for cases like McRae v CommonwealthDisposals Commission, where, in addition, recovery for the lost opportu-nity to profit from an alternative contract was allowed.92 Without endors-ing the extreme views put forward by Atiyah, the common law has stillfound its way around this seemingly ‘logical’ hurdle simply by being less‘dogmatic’ than the German civil law. Judges help out by skating over thisrough patch by stating that one who fails to perform his contract is justlybound to make good all damages that accrue or flow ‘naturally from thebreach’.93 In Anglia Television Ltd v Reed, Lord Denning simply askedwhether the loss—investments rendered futile by the breach—had beenwithin the contemplation of the parties and granted, without much ado,redress for the loss of investments incurred before the contract was evenentered into.94 Such rulings could be seen as the result of the judiciarybeing too little concerned with proper doctrinal analysis. However, these

88 H Altmeppen, ‘Untaugliche Regeln zum Vertrauensschaden und Erfüllungsinteresse imSchuldrechtsmodernisierungsentwurf’ [2001] Der Betrieb 1399, 1403ff; H Altmeppen,‘Nochmals: Schadensersatz wegen Pflichtverletzung, anfängliche Unmöglichkeit und Auf-wendungsersatz im Entwurf des Schuldrechtsmodernisierungsgesetzes’ [2001] Der Betrieb1821, 1823.

89 MB Kelly, ‘The Phantom Reliance Interest in Contract Damages’ [1992] WisconsinLaw Review 1755, 1775, n 60.

90 H McGregor, McGregor on Damages, 17th edn (London, Sweet & Maxwell, 2003)[2–020].

91 See, eg, EA Farnsworth, Contracts, 3rd edn (New York, Aspen Publishers, 1999) §12.16, 837 (‘Effect of losing contract’); JD McCamus, The Law of Contracts (Toronto, IrwinLaw, 2005) 835.

92 [1951] 84 CLR 377 (HCA).93 Security Stove & Mfg Co v American Rys Express Co 227 Mo App 175, 183 (1932),

citing Hobbs v Davis 30 Ga 423, 425 (1869).94 Anglia Television Ltd v Reed [1972] 1 QB 60 (CA).

46 Helge Dedek



rulings also show how the system leaves its judges room for the reasonableexercise of discretion—a system that has never truly believed in technicalperfection as the answer to the quest for justice.

Ironically—and counter-intuitively—it is the French civil law that hasmade the least efforts to lock in its judges by developing complicatedtheories around reliance and expectancy measure. It gets along withoutboth categories altogether, relying simply on the medieval distinctionbetween damnum emergens and lucrum cessans, the loss sustained and thegains prevented (see Article 1149 Code Civil (CC); Article 1611 Civil Codeof Quebec (CCQ)). Since the times of Cujas, out-of-pocket losses aresimply seen as damnum emergens:95 ‘Le calcul de la perte subie estrelativement facile à faire: ce sont les dépenses effectuées et devenuesinutiles.’96 The seemingly obvious lack of a causal nexus between breachand loss is not a matter of concern.97 The absence of the conceptualcategories (expectation/reliance damages) has been described in a way thatmakes it seem as if French law has simply not yet reached the level ofconceptual refinement to fully appreciate the distinction between expecta-tion and reliance interest.98 On the other hand, French civil law has notfaced the problems that arose from the conceptual petrifaction the Germansystem imposed on itself. Quite undogmatically, the French civilians simplyembraced the intuition that a breach ‘turns’ an investment made in relianceinto detriment, and therefore ‘causes’ a loss.

The idea that compensation for this ‘actual’ loss is an even more suitableand ‘natural’ reaction to the breach of promise was unfolded by Fuller andPerdue. By drawing on Aristotle, they demonstrated that compensation fora detrimental change of position in reliance on a promise is a matter ofcorrective justice and therefore represents a more pressing case for reliefthan the disappointment of having been deprived of the expectancy.99

German civil law, however, locked itself in by mechanically applying the

95 Iacobus Cuiacius, Opera Omnia In Decem Tomos Distributa, Opera Postumae QuaeDe Iure Reliquit, Tomus Quartus, Pars Prior, Paris, 1658, Comment. in Tit. I. de act.emp etvend. Lib. XIX Dig., ad l. 21, §.Cum per venditorem, Col 822: Dicimus autem circa remipsam, consistere pretium rei quod emptori abest, quod praenumeravit venditori, & etiamquod emptionis causa erogavit, puta, quod dedit parario, au courretier: quod dedit uxori (utfit nonnumquam) venditoris, quod dedit praeconi aut quaestori, quo interveniente factaauctio est, quod dedit fisco vectigalis nomine. [We, however, say that the interest relating tothe object of sale itself [circa rem ipsam] comprises the value of the thing the buyer has beendeprived of, the deposit the buyer gave to the seller, and also the expenses he made because ofthe contract of sale, eg, the amount he paid to the realtor, or the wife of the seller (as itsometimes happens), or to an auctioneer or a magistrate to avoid a sale by autction, the taxeshe paid to the authorities.]

96 Jobin, above n 65, at 905.97 G Ripert, Traité de droit civil: d’après le traité de Planiol (Paris, Librairie générale de

droit de jurisprudence, 1956) vol 2, 247.98 GH Treitel, Remedies for Breach of Contract (Oxford, Clarendon Press, 1988) 89.99 Fuller and Perdue, above n 76, at 56.




‘general rules on damages’. Ideas that do not fit the grid because theytranscend the ‘vertical’ borders have no place in legal discourse: thus the‘hardening of categories’ constricts the arsenal of legal reasoning.

IV. CONCLUSION

Observed from a comparative perspective, the attempts at ‘mapping’ thecommon law trigger ambivalent reactions. Particularly for civilians, noth-ing seems more alluring than a more coherent map of the common law100

which would help the outsider to overcome his confusion when trying tofind his way around. On the other hand, one is tempted to extend awarning: Be careful what you wish for! The civil law, and especially theGerman expression of the civilian tradition, has experienced the downsideof strict systematisation and classification. This temptation to warn mightalso have to do with the grass always seeming a bit greener on the otherside—while both civilians and common lawyers (more or less openly) tendto think of their respective ‘systems’ as somewhat superior (more orderly,more efficient, and so on), the harmony and order of the civil law does notfail to appeal to the common lawyer, whereas the freedom and flexibility ofcommon law reasoning tempts the civilian, in particular the academic.

I have emphasised, however, that it was not the purpose of this article tocriticise the intellectual merits of the ‘mapping’ enterprise. I simplytried—for purposes of comparison and, by all means, deterrence—tohighlight some problems that arise if a taxonomic approach to law ispushed to its extreme, as it has been in the German civil law tradition. Oneproblem arising in this context is that of excessive mechanical complica-tion. The German map of private law is so overly sophisticated that it, inall its complexity, really opens up only to the most ardent adept of German‘Legal Science.’ In a world where legal systems and styles are said to becompeting, this characteristic will—particularly in the light of the completeobliviousness of American law to such a kind of jurisprudence—be moreand more of a liability.

The foremost risk of taxonomy and classification in law, however, hasturned out to be the ‘hardening of categories’. Simply because taxonomy inlaw is not just descriptive, but prescriptive and normative in nature, onemust be always willing, once a classificatory scheme has been conceived, tocall it into question again. Stephen Smith acknowledges this necessity byintegrating into his twofold test the question of whether a classification

100 This yearning to understand and to appropriate also underlies the attempts ofcomparative functionalism to draw a global map and to develop trans-systemic categories ofcontractual concepts, as, eg, the famous category of the ‘Indicia of Seriousness,’ see Zweigertand Kötz, above n 40, at 388ff.

48 Helge Dedek



makes moral sense. This part of the test safeguards against the allure of a‘logical’ or ‘scientific’ classification that purports to be above and alooffrom current ideas and moralities.101 Becoming aware of this ‘moral’implication of legal taxonomy leads to the insight that we must not accepttraditional doctrinal categories as givens simply for their ‘logical’ appeal.On the other hand, if taxonomy has to account for its making ‘moral’sense, naturally the question arises, what does make ‘moral’ sense? Havingto answer this question as well makes the ambition of bringing a bit more‘logic’ to the common law, a truly Herculean task.102

Finally, the ‘hardening of categories’ once engrained might not just leadto the encapsulation of possibly outdated ideologies, but also, as I havetried to show, to an intellectual rigidity that curtails the argumentativepotential of legal discourse. The discourse censors ideas that do not fit thegrid of quasi-natural divisions and demarcation lines. However, everycartographer of knowledge must recall that ‘the map is not the terri-tory’.103 My conclusion would therefore be this. It could be reasonable andintellectually stimulating to redraw the map of contract law by establishing‘vertical borders’; however, keep such frontiers open so that ideas can stillmigrate freely. Beware of the petrification of categories. Classify with care;always taxonomise in moderation.

101 Which is, of course, even more problematic if such ideas have crystallised in thestructure of a Civil Code, which is, at least in France and Germany, seen as a nationaltreasure, and therefore tends to be fiercely guarded from change. On the attempts to initiate areform of the French Law of Obligations see, for an introduction, E Hondius, ‘The Two Facesof the Catala Project—Towards a New General Part of the French Law of Obligations’ (2007)15 European Review of Private Law 835–9.

102 See, eg, P Birks, ‘Equity in the Modern Law: An Exercise in Taxonomy’ (1996) 26University of Western Australia Law Review 1, 5.

103 A Korzybski, Science and Sanity: An Introduction to Non-Aristotelian Systems andGeneral Semantics, 3rd edn (Lakeville, Connecticut, The International Non-AristotelianLibrary, 1948) 58 (emphasis added).






3

Principle in Contract Law: theDoctrine of Consideration

STEPHEN WADDAMS

CHARLES ADDISON WROTE, in 1847, that

The law of contracts may justly indeed be said to be a universal law adapted toall times and races, and all places and circumstances, being founded upon thosegreat and fundamental principles of right and wrong deduced from naturalreason which are immutable and eternal.1

Many other nineteenth-century writers spoke also of the ‘principles’ ofcontract law, including Frederick Pollock (1876)2 and William Anson(1879),3 the titles of both of whose books commenced with the word‘Principles.’ These writers implied, both by the titles to their books and byremarks addressed to their readers, that there were certain propositionsabout English contract law, deserving of the name ‘principles’, that hadsome sort of special status as primary, fundamental, or indisputable, orthat constituted a source from which rules used to determine particularcases were derived, and that those propositions could be identified andformulated.

But if we ask precisely what were these principles, the answer provessurprisingly elusive. Part of the reason for this is the indeterminacy of theword ‘principle’: the word has been used in many different senses, themeaning varying according to what is implicitly contrasted with it (princi-ple and rule, principle and policy, principle and precedent, principle andauthority, principle and pragmatism, principle and practice, principle and

1 CG Addison, A treatise on the law of contracts and rights and liabilities ex contractu(London, W Benning, 1847) iv–v.

2 F Pollock, Principles of Contract at Law and in Equity (London, Stevens and Sons,1876).

3 W Anson, Principles of the English Law of Contract (Oxford, Clarendon Press, 1879).



convenience, principle and utility, general principle and particular rule,general principle and particular case); on a controversial legal question twoor more conflicting principles can usually—perhaps always—be identified;principles may be stated and restated at an infinite number of levels ofgenerality, and commonly the word has been used to mean no more than areason or rule framed at a higher level of generality than another.Sometimes a principle has meant more than a rule—a rule that isabsolutely stringent, but at other times the word has signified somethingless than a rule—an objective desirable in general terms but liable to beoutweighed by countervailing considerations. Whenever it is said—as itoften is—that two principles come into conflict, one or both of them areliable to be outweighed by countervailing considerations. Often the mean-ing of the word merges with the idea of ‘maxim.’ Very commonly also ithas signified a reasoned, or a well-reasoned legal argument; often it hasmeant a legal rule, or a reason in support of a rule, that the writer orspeaker considers persuasive, legitimate, or satisfactory.

As we have seen, Addison asserted in 1847 that the principles of contractlaw were immutable, eternal and universal. This kind of assertion rests ona ‘syncretic and ahistorical supposition’4 that all law everywhere must begoverned by the same principles. An examination of the doctrine ofconsideration, before and after 1847, shows that the conceptual bases ofthe doctrine, often, but not always, called principles, have varied markedlyfrom time to time, as has the substance of the law. A historical studycannot establish what is the correct or preferable meaning of the word‘principle’ nor what are the correct or preferable rules of contract law. Butit can show that neither the usage nor the law has been immutable oreternal.5

About 80 years before Addison’s statement William Blackstone hadpublished his Commentaries on the Laws of England. He likened his workto a map, writing that ‘an academical expounder of the laws . . . shouldconsider his course as a general map of the law, marking out the shape ofthe country, its connexions and boundaries, its greater divisions andprincipal cities.’6 Blackstone’s map gave no definite place to contract law,though he mentioned contracts in several places. A definition of contractwas offered in Book 2 (rights of things) as part of a chapter (entitled ‘Of

4 The phrase is David Ibbetson’s: Sir W Jones, An Essay on the Law of Bailments, editedwith introductions by David Ibbetson (Bangor, Welsh Legal History Society, 2004) 66.Syncretism, in theology, is the claim that all religions are fundamentally similar.

5 As for the ‘syncretic’ claim that the principles of contract law have been universal, thiswould have to be tested by a comparison of English law with other legal systems. It seemsimplausible that the claim could be sustained in respect of the doctrine of consideration,which is absent from many legal systems.

6 W Blackstone, Commentaries on the Laws of England (Oxford, Clarendon Press,1765) vol 1, 35.

52 Stephen Waddams



title by gift, grant, and contract’) dealing with methods of acquiring rightsto property.7 In Book 1 (rights of persons) contracts were mentioned aspart of the law of master and servant, and of husband and wife, and inBook 3 (private wrongs) assumpsit was mentioned as providing a remedyfor breaches of promises, considered as wrongs. In respect of husband andwife, Blackstone wrote that ‘our law considers marriage in no other lightthan as a civil contract’.8 This assertion is startling at first sight—so muchso that, when quoted, it has usually been to jocular or facetious effect, forthe differences, in Blackstone’s time, between marriage and other contractswere so many and so obvious that they scarcely require enumeration. Butthe question, for present purposes, is not what the assertion reveals aboutBlackstone’s view of marriage, but what it reveals about his view ofcontracts. The comment shows that Blackstone did not visualise contractsas a body of controlling principles from which legal obligations werederived: if he had thought in those terms he would have given contract lawa place on his map. It was not that Blackstone thought contractsunimportant—he refers to marriage as ‘the most important contract ofany’9—and contracts were an important means of transferring propertyrights, and an important aspect of the law of private wrongs. But he didnot think of legal issues as ‘part of’ an independently existing contract law.It would be more accurate to say that he thought of contracts as ‘part of’several different areas of the law—a means of effecting various kinds oflegal consequences—and hence to be found in several different places onhis map. The conception of contracts as a distinct body of principles wasnot, at this time, established in English law.

Often the idea of ‘principle’ has been used to denote the reason that issupposed to underlie a legal rule. One difficulty with this approach, from ahistorical perspective, is that it tends to dissolve the rule into the supposedprinciple. To take a simple example, the minimum age of contractualcapacity at common law was 21 years. It may be supposed that theunderlying reason for this rule was to ensure that contracting parties hadsufficient understanding and maturity. But a legal rule exists independentlyof the reason for it, even when a single reason can be identified, and, ofcourse, a rule that contracting parties must have attained the age of 21years is not the same thing as a rule that contracting parties must be ofsufficient understanding and maturity: there are many mature 20-year-olds,and many immature 21-year-olds. A more serious difficulty is that veryoften legal rules have been supported by several different reasons, whichmay be inconsistent with each other, and each one is insufficient, standingalone, to explain or support the rule. For example, two or more reasons

7 Ibid, at vol 2, ch 30.8 Ibid, at vol 1, 421.9 Ibid, at 424.

Principle in Contract Law 53



might be given in support of a legal rule, none of which is sufficient in itselfeither as explanation or as justification, and which, though often coincid-ing in their legal result, tend to require opposite results in particular cases.

Nowhere is this phenomenon more apparent than in relation to the rulethat a contractual promise must be supported by consideration. In anunpublished treatise written about 60 years earlier than Blackstone’sCommentaries, Sir Jeffrey Gilbert, later chief Baron of the Exchequer, likeBlackstone in his second book, envisaged contract primarily as a means oftransferring property. Gilbert commenced his treatise by saying:

Contracts are two-fold: verbal and solemn. Now contract is the act of two ormore persons concurring, the one in parting with, and the other in receivingsome property right or benefitt. The most notorious way of transferring of rightfrom one to the other is this by contract for all men by their labour and industrydid first acquire to themselves a property so they may by other acts of their owntransferr that property where they please, and all laws have allowed it as asettled maxim that the right of disposall must of necessity follow the rights ofabsolute dominium, for certainly as a man may be industrious for himself hemay be so also for another and therefore the establishment of the propriety [sic]must be in his hands to whom the disposition is made and no doubt as thenotion of propriety was begotten from humane necessity so was also this ofcontract.10

Gilbert then devoted many pages to the topic of consideration, with minutediscussion of numerous hypothetical and decided cases, showing thatconsideration was regarded by him as an important, difficult, controversialand complex topic. As his first reason for the legal requirement ofconsideration, Gilbert gives the need to protect potential defendants fromliability for rash promises. Having said that some opinions favoured ‘thepunctuall performance of every verbal promise’, he continued:

Others held that no obligation arises from a naked promise and that the force ofthe engagement doth totally depend on the consideration and they take it to be athing of great rigour that a man should dispose of the fruits and effects of a longand painfull industry and all the certain advantages and conveniences of life bythe meer breath of a word and the turn of an unwary expression; they also thinkthat the very laws of self-preservation will not permitt it for what reason ofconscience can oblige a man to those words that tend to his own destruction, butif a valuable consideration had been received the bargain is compleat for anotherman’s industry comes in the place of his own . . ..

Gilbert continued by saying that English law ‘hath held the middle betweenthese two extreames’, in that formal contracts were enforceable

10 J Gilbert, Of Contracts (c 1710) British Library, Hargrave 265, f 39 (some punctuationadded, abbreviations expanded, and capitalisation removed).

54 Stephen Waddams



so that if a man will oblige himself under the solemnitys of law whereby hiscontract appears to be seriously intended, it shall ever be obligatory and theconsideration shall be intended . . . but if the contract be verball only it binds inrespect of the consideration, otherwise a man might be drawn into an obligationwithout any real intention by random words, ludicrous expressions, and fromhence there would be a manifest inlet to perjury because nothing were more easythan to turn the kindness of expressions into the obligation of a real promise.11

‘Consideration’ is a conveniently flexible word, embracing three verydifferent ideas: deliberation, reason for making the contract, and reasonfor enforcing it. Gilbert’s reference to perjury adds a fourth idea: therequirement of consideration tends to supply reliable evidence that thepromise in question has actually been made.

This last idea was taken up by Lord Mansfield in Pillans and Rose v VanMierop and Hopkins.12 In that case a promise had been given in acommercial context by the defendant, in a signed writing, to guaranteerepayment of money already advanced by the plaintiff. Lord Mansfieldfavoured enforcement. He said:

I take it, that the ancient notion about the want of consideration was for thesake of evidence only; for when it is reduced into writing, as in covenants,specialties, bonds &c there was no objection to the want of consideration. Andthe Statute of Frauds proceeded upon the same principle. In commercial casesamongst merchants, the want of consideration is not an objection.13

Wilmot J thought that the requirement of consideration originally ‘wasintended as a guard against rash inconsiderate declarations’.14 This reasonalso would have been sufficient to justify enforcement in the particularcase, but he and the other judges managed, with some difficulty, to findconsideration (in the sense of value exchanged). Subsequent references tothe case have tended to amalgamate the principle favoured by LordMansfield (promises are enforceable if there is reliable evidence that theywere made) with that favoured by Wilmot J (promises are enforceable ifmade with due deliberation). The proposition that promises are enforce-able if made with serious intent, for example, tends to fuse the two ideas,and a requirement of writing might satisfy both simultaneously,15 but theideas are distinct as Lord Mansfield’s reference to the Statute of Fraudsreveals, for clearly there may be reliable evidence that a promise was in factmade, even though made rashly. Lord Mansfield, by appealing to aproposed principle, attempted to dissolve the rule that consideration was

11 Ibid, at 39–40ff (some punctuation added, abbreviations expanded, and capitalisationremoved).

12 (1765) 3 Burr 1663, 97 ER 1035.13 Ibid, at 1038.14 Ibid.15 See L Fuller, ‘Consideration and Form’ (1941) 41 Columbia Law Review 799.




always required into some such rule as that promises in writing, orcommercial promises in writing, were enforceable. Arguments can, ofcourse, be made in favour of Mansfield’s opinion, but it was not ahistorically accurate account of the previous law, and his view wasdecisively rejected by the House of Lords, on the advice of the judges, afew years later in Rann v Hughes.16 Standing alone, the need to protect thepromisor from rash promises could not explain the English law ofconsideration. Neither, standing alone, could the need for reliable evidence.But, as Gilbert’s writing shows, both these ideas have been influential.

Included in Gilbert’s account of consideration, quoted above, is anotherreason for the doctrine of consideration, namely, that it tends to preventthe dissipation of wealth by assuring to the promisor an equivalent inexchange for the wealth he gives up (‘what reason of conscience can obligea man to those words that tend to his own destruction, but if a valuableconsideration has been received the bargain is compleat for another man’sindustry comes in the place of his own’). Standing alone, this reason also isinsufficient to explain or to justify the actual law, because there was noneed in English law for consideration to be of equal value to the promisesought to be enforced. But this is not to deny that the reason was, as amatter of history, influential, or that, as a matter of fact, it did (in somecases, though not in all) tend to prevent the dissipation of wealth.

Another set of interrelated reasons for the doctrine of considerationrelates to the ideas of reciprocity and entitlement. As Guenter Treitelwrote, the claims of a promisee who has given nothing for the promise ‘areless compelling than those of a person who has given (or promised) somereturn for the promise’.17 One who has paid for a promise has a strongerclaim to assert an entitlement to performance of the promise than one whohas not paid for it, and the promisor has a correspondingly greaterobligation to perform the promise, and to perform it to its full extent. Oneof the reasons that tend to support specific performance and the expecta-tion measure of damages is that the promisee has bought and paid for theright to performance. Peter Benson has demonstrated the links among theconcept of entitlement, the doctrine of consideration, and the extent of theusual remedies for breach of contract.18

Yet another reason that has been given in support of the doctrine ofconsideration is that it tends to protect creditors in cases where thepromisor is insolvent. Lord Denman said, in Eastwood v Kenyon:

16 4 Bro PC 27, 2 ER 187. Summarised in (1778) 7 TR 350.17 GH Treitel, The Law of Contract, 11th edn (London, Sweet & Maxwell, 2003) 67.18 P Benson, ‘The Unity of Contract Law’ in P Benson (ed), The Theory of Contract Law

(Cambridge, Cambridge University Press, 2001) 118.

56 Stephen Waddams



The enforcement of such promises by law, however plausibly reconciled by thedesire to effect all conscientious engagements, might be attended with mischie-vous consequences to society; one of which would be the frequent preference ofvoluntary undertakings to claims for just debts. Suits would thereby be multi-plied, and voluntary undertakings would also be multiplied, to the prejudice ofreal creditors. The temptations of executors would be much increased by theprevalence of such a doctrine and the faithful discharge of their duty be renderedmore difficult.19

Where, as is common in cases of charitable pledges, the action is broughtagainst the estate of the promisor after death, there is the added considera-tion that enforcement of the promise will be at the expense of possiblyneedy dependants.20

One reason commonly given for the enforcement of promises is that ittends to protect the reliance and expectation of the promisee. If, as justsuggested, one who has paid for a promise has a stronger claim to enforceperformance than one who has not paid, enforcement and considerationare necessarily interlinked. William Paley, in his Principles of Moral andPolitical Philosophy (1785) gave this as the primary reason for enforcingpromises:

Those who argue from innate moral principles suppose a sense of the obligationof promises to be one of them; but, without assuming this, or anything else,without proof, the obligation to perform promises may be deduced from thenecessity of such a conduct to the well-being, or the existence indeed, of humansociety. Men act from their expectations. These expectations are in most casesdetermined by the assurances and engagements we receive from others. If nodependence could be placed upon these assurances, it would be impossible toknow what judgment to form of many future events, or how to regulate ourconduct with respect to them. Confidence therefore in promises, is essential tothe intercourse of human life . . . But there could be no confidence in promises, ifmen were not obliged to perform them: the obligation therefore to performpromises is essential, to the same end, and in the same degree.21

Paley was not a lawyer, but he had a strong influence on nineteenth-century English thought, including legal thought. His Principles was formany years included in the very select reading list of the Cambridge Boardof Moral Sciences Studies.22 A long extract from his chapter on promises,with an elaborate example designed to show that promises were to be

19 Eastwood v Kenyon (1840) 11 Ad & E 438, 113 ER 482, 487.20 In the leading Canadian case of Governors of Dalhousie College v Boutilier Estate

[1934] SCR 642, 3 DLR 593 the promisor had suffered ‘severe financial reverses whichprevented him from honouring his pledge’. The action was brought against the estate after thedeath of the promisor.

21 W Paley, Principles of Moral and Political Philosophy (London, 1785) 106.22 W Whewell, Lectures on the History of Moral Philosophy in England, new edn

(Cambridge, Deighton Bell, 1862) 279.




interpreted in the sense understood by the promisee, if known to thepromisor, and not in the sense subjectively intended by the promisor, wasincluded in Joseph Chitty’s much-used treatise on the law of contracts (2ndedition, 1834) and maintained in successive editions until nearly the end ofthe century.23 Paley was expressly relied on in argument in the importantcase of Smith v Hughes,24 and plainly influenced the court’s formulation inthat case of the objective principle of contractual obligations.25

Paley continued his discussion of promises by considering and rejectingthe argument that society could manage satisfactorily without enforceabil-ity of promises:

Some may imagine, that, if this obligation were suspended, a general caution,and mutual distrust would ensue, which might do as well: but this is imagined,without considering, how every hour of our lives we trust to and, depend uponothers; and how impossible it is, to stir a step, or, what is worse, to sit still amoment without such trust and dependance. I am now writing at my ease, notdoubting (or rather never distrusting, and therefore never thinking about it) butthat the butcher will send in the joint of meat, which I ordered; that his servantwill bring it; that my cook will dress it; that my footman will serve it up; andthat I shall find it upon [my] table at one o’clock. Yet have I nothing for all this,but the promise of the butcher, and the implied promise of his servant and mine.And the same holds, of the most important, as well as the most familiaroccurrences of social life.26

It is notable that all the examples given are of exchange transactions. If onethinks, as Paley did, of what promises are useful or necessary to anorganised society, one thinks naturally of exchange transactions, not of giftpromises. It is easy to imagine a well-organised society in which giftpromises are not enforceable—eighteenth-century English society was, ofcourse, a case in point—but it is more difficult to imagine an organisedsociety without enforcement of exchange transactions. In this way, consid-eration has been linked with ideas of expectation and reliance, and, moregenerally, of social order.

A rule supported by so many and so various reasons leads inevitably toinstances where the rule, as currently formulated, applies, but where the

23 J Chitty, A Practical Treatise on the Law of Contracts not under Seal: and upon theusual Defences to Actions thereon, 2nd edn (London, S Sweet, 1834) 62, 12th edn (London,Sweet & Maxwell, 1890) 127–8. Not found in 1st edn (1826) or in 13th edn (1896).

24 (1871) LR 6 QB 597.25 Ibid. Blackburn J said, at 607, ‘If, whatever a man’s real intention may be, he so

conducts himself that a reasonable man would believe that he was assenting to the termsproposed by the other party, and that other party upon that belief enters into the contractwith him, the man thus conducting himself would be equally bound as if he had intended toagree to the other party’s terms.’ Paley was expressly mentioned by Hannen J, at 610, for theimportant corollary that a promisor cannot be held to a meaning known by the promisee notto have been intended.

26 Paley, above n 21, at 106–7.

58 Stephen Waddams



underlying reasons, or some of them, do not. English law has had muchdifficulty with the question of modification of contracts, where additionalvalue is promised by one party in exchange for performance by the other ofan obligation already due under an earlier contract. It has often occurredthat during performance of a contract circumstances change so that a partyis in a position to demand from the other a higher payment than thatoriginally agreed. A number of nineteenth-century cases involved sailors,who, having agreed to serve for the whole of a voyage at a certain wage,subsequently demanded a higher wage when the ship was at a place wheresubstitute services were not readily obtainable. The renegotiated contractswere generally set aside. According to Campbell’s report of the leading caseof Stilk v Myrick, the reason for this result was that performance of apre-existing contractual duty could not constitute consideration.27 Thus,where a contract had been entered into at a fixed price, a subsequentrenegotiation or variation consisting of an agreed increase in the price wasunenforceable.

The rule in Stilk v Myrick (as reported by Campbell, and as it wasgenerally understood to be) was much criticised on the grounds that it didnot correspond to commercial understanding, that it failed to recognisethat actual performance was of greater real practical value than a legalright to performance, and that it was easily circumvented by the parties orby a court desirous of enforcing the variation.28 But criticism was oftentempered with the observation that the rule, though difficult to defend interms of consideration, was yet serving a useful purpose in offering, albeitindirectly, some legal protection against taking undue advantage of eco-nomic pressure. Stilk v Myrick itself was a case in point, where sailors,having agreed to serve on a voyage for certain wages, were promisedhigher wages in order to induce them not to desert during the course of thevoyage. It is evident that one of the reasons for the decision was to protectthe shipowner from a potentially extortionate threat by the crew to desertthe ship in a distant place where there was no ready supply of substitutelabour. In an earlier case, Harris v Watson, the court had also refused toenforce such a contract, but had given as the reason that

if this action was to be supported, it would materially affect the navigation ofthis kingdom . . . for if sailors were in all events to have their wages, and in timesof danger entitled to insist on an extra charge on such a promise as this, theywould in many cases suffer a ship to sink, unless the captain would pay anyextravagant demand they might think proper to make.29

27 (1809) 2 Camp 317, 170 ER 1168.28 B Reiter, ‘Courts, Consideration, and Common Sense’ (1977) 27 University of Toronto

Law Journal 439.29 (1791) Peake 102, 170 ER 94.




A different report (Espinasse) of Stilk v Myrick, running together the ideasof principle and policy, states that the judge (Lord Ellenborough) ‘recog-nised the principle of the case of Harris v Watson as founded on just andproper policy’,30 and in a subsequent decision in the Admiralty Court (TheAraminta, 1854) where the sailors had secured payment in gold of theextra money in advance of the extra performance (they were tempted todesert to gold diggings in Australia in 1852) and so the doctrine ofconsideration was of no assistance, Stilk v Myrick was interpreted asholding that the variation of the contract was ‘illegal’.31

In Williams v Roffey Bros & Nicholls (Contractors) Ltd (1990) theEnglish Court of Appeal held a renegotiation to be enforceable.32 In thatcase, a subcontractor had contracted to perform carpentry work at anagreed price. When the work was partly done it became clear that thesubcontractor would not complete it at the contract price, and the headcontractor, who was subject to a penalty clause in the main contract fordelay in completion, agreed to pay a higher price for the carpentry work.This latter agreement was held to be enforceable. The court held thatperformance of an existing obligation might constitute consideration.References to ‘principle’ were prominent. Glidewell LJ, who gave theleading judgment, rejected the argument that this conclusion was contraryto principle:

If it be objected that the propositions above contravene the principle in Stilk vMyrick I answer that in my view they do not; they refine, and limit theapplication of that principle, but they leave the principle unscathed . . . it is notin my view surprising that a principle enunciated in relation to the rigours ofseafaring life during the Napoleonic wars should be subjected during thesucceeding 180 years to a process of refinement and limitation in its applicationin the present day.33

But Glidewell LJ added the very significant proviso that the renegotiationwould be liable to be set aside if there was economic duress, which hecalled ‘another legal concept of relatively recent development’,34 therebysuggesting that the result in Stilk v Myrick might be supported, though noton the reasoning given in Campbell’s report. Many have welcomed thedemise of consideration in this context, but it is not easy to say preciselywhat has replaced it. What exactly, in the court’s view, was the governingconcept of enforceability and how did it apply in practice to contractual

30 6 Esp 129, 130, 170 ER 851. Espinasse did not have a very high reputation as areporter, but on the other hand he was one of the counsel in the case and had the means ofknowledge.

31 The Araminta, (1854) 1 Sp Ecc & Ad 224, 164 ER 130 (Adm Ct).32 [1991] 1 QB 1 (CA) 10, 19.33 Ibid, at 16.34 Ibid, at 13.

60 Stephen Waddams



renegotiations? These questions are not very easily answered.35 Oneapproach has been to attempt to distinguish between a ‘threat’ and a mere‘offer to renegotiate’, on the ground that there is something wrongfulabout a threat but not about an offer.36 In Williams v Roffey Bros it wassuggested that there was no duress because the proposal for renegotiationemanated from the head contractor. Purchas LJ thought this a conclusivepoint,37 but it is evident that the subcontractor had indicated, by conduct ifnot by words, that he was not likely to complete the work on time at thecontract price. It cannot be crucial that the threat not to complete wasimplicit rather than explicit. Indeed, it may be in the very cases where thereis no real choice that it is unnecessary to spell out the threat, or even tomake what could readily be called a ‘threat’ at all. In The Araminta, thecase of the payment to the ship’s crew at the Australian gold diggings in1852, it was the master who, after several desertions, took the initiativeand called together the rest of the crew, offering them increased wages ifthey would work the ship short-handed. Dr. Lushington said, of themaster’s payment, that it was made voluntarily, adding:

I have used the expression voluntarily, because I think the effect of the evidenceis, that the crew exercised no compulsion towards him, though, perhaps, inanother sense of the word, such payment was not voluntary, and the more aptexpression may be, and the one nearest the truth, that he was compelled bycircumstances to make that payment.38

This is indeed often an apt expression to describe such circumstances, andfor this reason it is doubtful whether the conclusion can be resisted that inWilliams v Roffey Bros, as in most cases of this sort, there was athreatened breach of the first contract. The decision that the modifiedcontract was nevertheless enforceable lends support to the view that, wherethe pressure on the other party is not excessive,39 many courts haveaccepted that it is legitimate to gain an advantage in this way.

This leaves it very difficult to state what principles govern modificationof contracts. Glidewell LJ said that the principle of Stilk v Myrick was left‘unscathed’,40 but any formulation of that principle before 1990 would

35 For a fuller discussion see S Waddams, ‘Commentary on “The Renegotiation ofContracts”’ (1998) 13 Journal of Contract Law 199; S Waddams, ‘Unconscionable Contracts:Competing Perspectives’ (1999) 62 Saskatchewan Law Review 1.

36 See S Smith, ‘Contracting under Pressure: a Theory of Duress’ [1997] CLJ 343.37 Williams v Roffey Bros, above n 32, at 21.38 The Araminta above n 31, at 133 (emphasis in original).39 Restatement (Second) of Contracts (1979) § 89 provides that modifications are

enforceable if ‘fair and equitable’ in view of unexpected circumstances. The UniformCommercial Code, s 2–209 makes modifications enforceable, explicitly subject however, bycomment 2, to a test of good faith.

40 Purchas LJ said, at 21, that ‘the rule in Stilk v Myrick remains valid as a matter ofprinciple, namely, that a contract not under seal must be supported by consideration’, but the




have plainly demanded the opposite result to that reached in Williams vRoffey Bros.41 Glidewell LJ was consciously changing the law as it hadpreviously been understood to be, but he found it necessary to say that the‘principle’ remained intact. Though ‘unscathed’ (that is to say, undamaged)the principle is said to be refined and limited, but it is not at all clear whatthese refinements and limitations amount to. Consideration is still neces-sary; however, performance of a pre-existing contractual duty may amountto consideration (but does not always do so); however, again, even if thereis consideration the modification will not be enforceable if there iseconomic duress; however, yet again, it is not explained why the quitesevere pressure on the head contractor in Williams v Roffey Bros. (thethreat of the penalty clause) did not amount to duress. Breach of contractis a wrong, but apparently it is permissible to gain an advantage bythreatening to break a contract. Economic duress was said, by GlidewellLJ, to ‘provide another answer in law to the question of policy which hastroubled the courts since before Stilk v Myrick and no doubt led, at thedate of that decision, to a rigid adherence to the doctrine of considera-tion’.42 ‘Rigid,’ in legal argument, is never a word of approbation, so thesuggestion seems to be that Stilk v Myrick can be justified only on the basisof ‘policy’, but it is not very clear what the policy was or is (protection ofthe navigation of the Kingdom, or avoidance of economic duress), or howit relates to the ‘unscathed’ principle.

Another aspect of the doctrine of consideration has been the trouble-some question of contracts for the benefit of third parties. In the old caseof Dutton v Poole, a father, wishing to give money to his daughter andproposing to cut down trees to raise the money, agreed with his son andheir that he would refrain from cutting down the trees if the son would paythe daughter £1,000.43 The son inherited the land with the timber intact,but refused to honour his promise. The promise was held to be enforce-able.

Until the mid-nineteenth century it was generally accepted by writersthat this case was good law, though it was evidently an exception to the

proposition that a contract must be supported by consideration is not the ‘principle’ or the‘rule’ for which Stilk v Myrick had been cited during the previous 180 years; Purchas LJ seemsto be suggesting, in view of the conclusion in Williams v Roffey Bros, that, though the ‘rule’in Stilk v Myrick ‘remains valid as a matter of principle’, the principle must have beenmisapplied in Stilk v Myrick itself.

41 For example, GC Cheshire and CHS Fifoot, The Law of Contract, 6th edn (London,Butterworths, 1964) 77, calling it ‘the somewhat obvious rule, that there is no considerationif all that the plaintiff does is to perform or to promise the performance of an obligationalready imposed upon him by a previous contract between him and the defendant’.

42 Above n 32, at 14.43 (1689) 2 Lev 210, 83 ER 523, affirmed T Raym 302, 83 ER 156 (Ex Ch).

62 Stephen Waddams



idea of privity of contract.44 A variety of explanations was proposed.Gilbert approved of the result of the case, but encountered some difficultywith formulating a principle and defining its limits:

So in consideration that the father of the defendant whose heir he is would notcut trees to pay the portion of the wife of the plaintiff sister of the defendant heassumed to pay to the wife of the plaintiff 1000l; the daughter and her husbandmay have an action on her father by reason of the nearness of relation and it is adebt to the daughter to make a provision [portion?] for her and so she isinterested in the consideration and the son hath benefit by it; otherwise it is if themoney had been paid to a stranger; the law therefore will put a near relation inthe place of the promisee because the promisee is bound by the law of nature toprovide for such relation and therefore the labour of the relation is a considera-tion for the promise made to him and the relation hath loss by not receiving thevalue of his father’s labour but the law will not put a stranger in the place of thepromisee, because there is no consideration why the payment shall be made to astranger from whom the consideration did not rise.45

This laborious reasoning is not very convincing to the modern reader, andit is of interest in the present context for precisely that reason. Gilbert feltthe necessity of formulating a proposition that would explain and justifythe decision in Dutton v Poole, while maintaining a general rule of privityof contract and excluding the ‘stranger.’ Several possible propositions aresuggested by this passage: the promise is enforceable where promisee andbeneficiary are close relatives; the promise is enforceable where thepromisee owes a moral duty to confer the benefit on the beneficiary; thepromise is enforceable where some asset of the promisee that wouldotherwise have been transferred to the beneficiary is transferred instead tothe promisor in exchange for the promise. All these formulations havedifficulties, and it is not the object of the present inquiry to consider whichis preferable, or which most accurately represented the law in 1678 or inGilbert’s own time. The significant point is that Gilbert felt the need toarticulate a proposition, or principle, that would satisfactorily explain andjustify Dutton v Poole, while at the same time stating the law as it wasperceived to be when he wrote, and laying down a rule that would (in hisand his readers’ opinion) be satisfactory for the disposition of future cases.

The reason for the conclusion in Dutton v Poole plainly had somethingto do with general considerations of justice, including unjust enrichment.As the report puts it, ‘the son hath the benefit by having of the wood, and

44 S Comyn, The Law of Contracts (1826) vol 1, 27; Addison, above n 1, at 247–8,quoting Lord Mansfield in Martyn v Hind (1779) 1 Doug 142, 99 ER 94, 96: ‘it is difficult toconceive . . . how a doubt could have been entertained in the case of Dutton v Poole’; Chitty,A Treatise upon the Law of Contracts, 2nd edn above n 23, at 48 (though with somereservation).

45 Gilbert, above n 10, at n 82 (some punctuation added, abbreviations expanded, andcapitalisation removed).




the daughter hath lost her portion by this means’.46 Of course, the phrase‘unjust enrichment’ was not in use in the seventeenth century, but plainerlanguage could scarcely have been found to express the idea that the sonhad been unjustly enriched at the expense of the daughter. Gilbert also, inthe passage just quoted, mentions that ‘the son hath benefit by it’, andmakes the argument that the daughter suffers a loss by the transfer of thevalue of her father’s ‘labour’, (that is, a valuable asset, in this case thetimber), to the son instead of to her.

In the nineteenth century, Dutton v Poole came to seem inconsistentwith the principle of consideration, and the case was rejected in Tweddle vAtkinson.47 Crompton J expressly recognised that the law had changed:

At the time when the cases which have been cited [these included Dutton vPoole] were decided the action of assumpsit was treated as an action of trespassupon the case, and therefore in the nature of a tort; and the law was not settled,as it is now, that natural love and affection is not a sufficient consideration for apromise upon which an action may be maintained; nor was it settled that thepromisee cannot bring an action unless the consideration for the promise movedfrom him. The modern cases have, in effect, overruled the old decisions; theyshew that the consideration must move from the party entitled to sue upon thecontract. . . . I am prepared to overrule the old decisions, and to hold that, byreason of the principles which now govern the action of assumpsit, the presentaction is not maintainable.48

This is a very revealing passage, and shows the self-consciously novel viewof principle that came to predominate in the mid-nineteenth century. The‘principles which now govern the action of assumpsit’ required theoverruling of ‘the old decisions’, even though no one doubted the justice(as between the parties) of Dutton v Poole, the case had stood for nearly200 years, and the Queen’s Bench had no actual power to overrule it.49

Blackstone’s treatment of assumpsit as part of private wrongs, less that 100years old, was summarily consigned to a primordial period of unprincipledignorance, which prevailed until the law was ‘settled, as it is now’. Theselast words reflect the Victorian confidence in the perfectability, andnearly-attained perfection, of English social institutions, including law.

The approach in Tweddle v Atkinson was confirmed by the House ofLords in 1915. The concept of principle was again prominent. ViscountHaldane said that:

in the law of England certain principles are fundamental. One is that only aperson who is a party to a contract can sue on it. Our law knows nothing of ajus quaesitum tertio arising by way of contract. . . . A second principle is that if

46 Above n 43, at 524.47 (1861) 1 B & S 393, 121 ER 762.48 Ibid, at 764.49 As pointed out by Blackburn J, ibid.

64 Stephen Waddams



a person with whom a contract not under seal has been made is to be able toenforce it consideration must have been given by him to the promisor or to someother person at the promisor’s request. These two principles are not recognisedin the same fashion by the jurisprudence of certain Continental countries or ofScotland, but here they are well established.50

The link between this conclusion and the doctrine of consideration wasemphasised by Lord Dunedin, who, comparing Scots law, to its advantage,with English law, said pointedly:

I confess that this case is to my mind apt to nip any budding affection which onemight have had for the doctrine of consideration. For the effect of that doctrinein the present case is to make it possible for a person to snap his fingers at abargain deliberately made, a bargain not in itself unfair, and which the personseeking to enforce it has a legitimate interest to enforce. Notwithstanding theseconsiderations I cannot say that I ever had any doubt that the judgment of theCourt of Appeal was right [as to English law].51

Lord Haldane’s assertion that privity of contract was a ‘fundamentalprinciple’ was challenged by Denning LJ, who said, in 1949,

Counsel . . . says that the plaintiffs cannot sue. He says that there is no privity ofcontract between them and the board, and that it is a fundamental principle thatno one can sue on a contract to which he is not a party. That argument can bemet either by admitting the principle and saying that it does not apply to thiscase, or by disputing the principle itself. I make so bold as to dispute it. Theprinciple is not nearly so fundamental as it is sometimes supposed to be. It didnot become rooted in our law until the year 1861 . . . and reached its full growthin 1915.52

However, in 1962 the principle was reasserted by the House of Lords (LordDenning, then himself a member of the House of Lords, dissenting).Viscount Simonds said

Learned counsel . . . met [the argument for enforcement] . . . by asserting aprinciple which is, I suppose, as well established as any in our law, a ‘fundamen-tal’ principle, as Lord Haldane called it, . . . an ‘elementary’ principle, as it hasbeen called times without number, that only a person who is a party to acontract can sue upon it. ‘Our law’, said Lord Haldane, ‘knows nothing of a jusquaesitum tertio arising by way of contract’. Learned counsel . . . claimed thatthis was the orthodox view and asked your Lordships to reject any propositionthat impinged upon it. To that invitation I readily respond. For to me hetero-doxy, or, as some might say, heresy, is not the more attractive because it isdignified by the name of reform. Nor will I easily be led by an undiscerning zealfor some abstract kind of justice to ignore our first duty, which is to administer

50 Dunlop Pneumatic Tyre Co Ltd v Selfridge & Co Ltd [1915] 1 AC 847 (HL) 853.51 Ibid, at 855.52 Smith v River Douglas Catchment Board [1949] 2 KB 500 (CA) 514.




justice according to law . . . The law is developed by the application of oldprinciples to new circumstances. Therein lies its genius. Its reform by theabrogation of those principles is the task not of courts of law but of Parliament[two decisions of Lord Denning were cited here and rejected].53

This exceeds the forceful, and presses the limits of judicial courtesy. Onecould scarcely imagine a more powerful assertion of the absolute immuta-bility of legal principle. Yet, surprising as this may seem, it was in effectabandoned within a few years, when the opposite result was reached by thePrivy Council on similar facts in the New Zealand case of The Euryme-don.54 Lord Wilberforce, giving the judgment of the majority of theJudicial Committee, said that the conclusion (enforcement by the thirdparty) could ‘be given within existing principles’, but he had considerabledifficulty in explaining the result in conventional terms. The Board washeavily influenced by considerations of commercial convenience and gen-eral considerations of fairness. Referring to an American case, LordWilberforce said that ‘commercial considerations should have the sameforce on both sides of the Pacific’, and said that he desired ‘to give effect tothe clear intentions of a commercial document’ adding:

It should not be overlooked that the effect of denying validity to the clausewould be to encourage actions against servants, agents and independent contrac-tors in order to get round exemptions (which are almost invariable and oftencompulsory) accepted by shippers against carriers, the existence, and presumedefficacy, of which is reflected in rates of freight. They see no attraction in thisconsequence.55

This was a clear victory for considerations of commercial convenience andgeneral considerations of justice, and an effective abandonment of whathad so recently been asserted by the House of Lords as ‘fundamentalprinciple’.

In a modern case having some parallels with the old case of Dutton vPoole, namely, Beswick v Beswick, an uncle transferred a coal business (hisonly substantial asset) to his nephew in exchange for the nephew’s promiseto pay an annuity to the uncle’s widow.56 The case resembles Dutton vPoole in that the promisor had actually benefited by receiving a valuableasset that would otherwise have benefited the plaintiff (because the unclewould otherwise have made other provision for her), differing in thisrespect from Tweddle v Atkinson, where no payments had been made. AsGilbert might aptly have said, the nephew had benefited by receiving thevalue of his uncle’s labour, and the widow had lost correspondingly. But

53 Scruttons Ltd v Midland Silicones Ltd [1962] 1 AC 446 (HL) 467–8.54 New Zealand Shipping Co Ltd v A M Satterthwaite & Co Ltd [1975] 1 AC 154 (PC)

169 [The Eurymedon].55 Ibid, at 169.56 [1968] AC 58 (HL).

66 Stephen Waddams



this feature of the cases, of such obvious importance to every considerationof justice between the parties, was made to appear irrelevant by theprinciple established in Tweddle v Atkinson combined with an over-rigidscheme of categorisation that excluded considerations of unjust enrich-ment.57 It is little to the credit of the law to establish, in the name of‘principle’, rules that require the court to close its eyes to factors crucial tothe attainment of justice.

The promise in Beswick v Beswick was held to be enforceable, but onlybecause the widow happened to be the administratrix of the uncle’s estate,and so entitled, in the opinion of the House of Lords, to a decree of specificperformance. It is plain that the court was influenced by the generalconsiderations of justice just mentioned. The law lords described thenephew’s conduct as ‘an unconscionable breach of faith’,58 and thepossibility of there being no remedy as ‘grossly unjust’,59 and ‘repugnant tojustice and [such as to] fulfil no other object than that of aiding thewrongdoer’.60 The use of specific performance for this purpose was highlyunusual. The remedy was given not, of course, because there was anythingin the nature of money that could not compensate, but in order tocircumvent the rule of privity, which Lord Pearce called ‘a mechanicaldefect of our law’.61 Lord Reid indicated, contrary to Viscount Simonds’view, that reform of the law would, if necessary, be within the properpower of the court.62 Statutory reform of English law had been recom-mended by the Law Revision Committee in 1937, but nothing was done byParliament until the enactment of the Contracts (Rights of Third Parties)Act, 1999.

In Canada, the English law of the late nineteenth and twentieth centurieswas strictly followed, leading to a result in Greenwood Shopping Plaza Ltdv Beattie63 which can, without much exaggeration, be called absurd.64 Thelessor of business premises had covenanted with its tenant to insure againstfire. A loss by fire occurred, allegedly caused by the negligence of two ofthe tenant’s employees, and the Supreme Court of Canada held that thelessor (and its insurer) was entitled to sue the two employees individuallyfor the whole of the loss, even if the proper interpretation of the contractwas that it had promised not to do so. The Supreme Court of Canada said

57 See S Waddams, Dimensions of Private Law: Categories and Concepts in Anglo-American Legal Reasoning (Cambridge, Cambridge University Press, 2003).

58 Lord Hodson, above n 56, at 83.59 Lord Reid, ibid, at 73.60 Lord Pearce, ibid, at 89.61 Ibid.62 Ibid, at 72. To similar effect Lord Scarman in Woodar Investment Development Ltd v

Wimpey Construction UK Ltd [1980] 1 WLR 277 (HL) 300.63 [1980] 2 SCR 228, 111 DLR (3d) 257.64 See below n 80.




that ‘the rule of privity . . . since Tweddle v Atkinson . . . has had decisiveeffect in this branch of the law. There are many cases which have appliedthis principle.’65 As the Chief Justice of the Nova Scotia Court of Appealhad pointedly commented, this result ‘[flew] in the face of common sense,modern commercial practice and labour relations’.66

When the issue arose again in London Drugs Ltd v Kuehne & NagelInternational Ltd the Supreme Court of Canada took a very differentview.67 The facts were rather similar to those in Greenwood. The plaintiffstored a valuable transformer with the defendant warehouser, agreeing tolimit liability to $40. The transformer was damaged by the negligence oftwo employees, and, as in Greenwood, the owner sued the employeespersonally. The plaintiff’s counsel relied on ‘longstanding, established andfundamental principles of law’, no doubt with some confidence of successin view of the quite recent decision of the court in Greenwood. However,Iacobucci J, giving the judgment of the majority of the court, decided infavour of the employees. Iacobucci J could easily have found that the casefell into one of the established exceptions to the doctrine of privity, but hechose instead to deal with the issue directly, saying ‘I prefer to deal head-onwith the doctrine of privity and to relax its ambit in the circumstances ofthis case’.68 He considered that the strict rule should be relaxed for reasonsof ‘commercial reality and common sense’.69 Similar expressions wererepeated: ‘sound commercial practice and justice’,70 ‘the reasonable expec-tations of all the parties to the transaction’,71 ‘the underlying concerns ofcommercial reality and justice’,72 ‘commercial reality’,73 a result that made‘sense in the modern world’,74 ‘sound policy reasons’,75 ‘commercial realityand justice’,76 and ‘modern notions of commercial reality and justice’,77

ideas that were contrasted, to their advantage, with ‘a strict application ofthe doctrine of privity’,78 and ‘the rigid retention of a doctrine that hasundergone systematic and substantial attack’.79 He said that it would be

65 Above n 63, at 263.66 Greenwood Shopping Plaza Ltd v Neil J Buchanan Ltd (1979) 99 DLR (3d) 289

(NSAD) 295 (MacKeigan CJNS).67 [1992] 3 SCR 299, 97 DLR (4th) 261.68 Ibid, at 341 (DLR).69 Ibid, at 342.70 Ibid, at 348.71 Ibid.72 Ibid.73 Ibid, at 360.74 Ibid, at 364.75 Ibid.76 Ibid, at 365.77 Ibid, at 370.78 Ibid, at 361.79 Ibid, at 358.

68 Stephen Waddams



‘absurd in the circumstances of this case to let the appellant go around thelimitation of liability clause by suing the respondent employees in tort’.80

The court did not, however, simply abolish the doctrine of privity, butcreated a limited exception, which Iacobucci J described as an ‘incrementalchange’.81 Many commentators and courts had said, as we have seen, thatit was a ‘principle’—often indeed called a ‘fundamental principle’—ofEnglish and Canadian law that only a party to a contract could sue on it.Iacobucci J himself described privity of contract as ‘an established principlein the law of contracts [which] should not be discarded lightly’.82 Thisphrase demonstrates the elusive meaning of the idea of principle. Evidentlya ‘principle,’ even though ‘established’, is not necessarily determinative oflegal issues, but can be ‘discarded’ for sufficient reason (though not‘lightly’). Although in that sentence Iacobucci J called privity of contract a‘principle’, more often he described privity as a ‘doctrine’ and the excep-tion to it as ‘principled’.83 Some might wish that the court had made amore radical change. On this it may be remarked first that an exception onsuch general grounds as ‘commercial reality and justice’ is in reality a veryfar-reaching change, and second that there are good reasons for caution. Asimple declaration by the court that the rule of privity was abolished wouldhave compelled future courts to enforce contracts for the benefit of thirdparties; the recognition of a limited exception, on the other hand, has theeffect of empowering future courts in appropriate cases to enforce suchcontracts, but not compelling them to do so. There are good reasons not tolay down a rule that all contracts for the benefit of third parties mustalways be enforced. There are two particularly difficult cases. One is thecase of the incidental beneficiary—one who would have benefited byperformance of the contract but not a person on whom the contractingparties intended to confer rights. The other difficult case is where theoriginal contracting parties seek to rescind or modify the contract. Some-times it is appropriate for them to do so, and sometimes it is moreappropriate to require the consent of the third party, but it is not easy toformulate a universal rule on the point. The effect of the decision inLondon Drugs was to reintroduce flexibility to the common law, and toenable the lower courts to reach fair and just results. This approach hasadvantages over statutory reform, which, as the English experience shows,is apt to lead to unexpected anomalies and complexities.84 In the Canadian

80 Ibid, at 363.81 Ibid, at 366.82 Ibid, at 358.83 See below n 86.84 The compexities of the English statute are forcefully demonstrated by R Stevens, ‘The

Contracts (Rights of Third Parties) Act, 1999’ (2004) 120 LQR 292.




context there is the added point that uniform provincial legislation on thematter could not realistically have been anticipated.

In the subsequent case of Fraser River Pile & Dredge Ltd v Can-DiveServices Ltd Iacobucci J, giving the judgment of the whole court, extendedthe London Drugs case to a case involving waiver by an insurer ofsubrogation rights.85 The decision shows that the recognition of third partyrights in contracts is not limited to any particular class of contract, andthere seems no reason why third party rights should not be recognised inany case where considerations of justice require it. Iacobucci J describedthe London Drugs case as having introduced a ‘principled exception to thecommon law doctrine of privity of contract’.86 Principle and policy wereclosely associated in his mind. In Fraser River he speaks of ‘policy reasonsin favour of an exception’,87 and, as in London Drugs of ‘common senseand commercial reality’.88 Sometimes a distinction has been made betweenprinciple and policy, but it is evident that in Iacobucci J’s mind they werenot opposed: on the contrary, good policy was an essential aspect of soundprinciple. The same could be said of many influential judges and commen-tators in the nineteenth and twentieth centuries.

Perhaps the greatest difficulty posed by the English doctrine of consid-eration has been the treatment of reliance. Subsequent reliance on apromise does not constitute consideration, but the consequence of refusingenforcement can be severe injustice. One of the simplest imaginableexamples of reliance arises when a landowner promises to give the land toanother person (for example, a relative) and the other person, relying onthe promise, builds on the land. The promisor (or, as it has more usuallybeen, his or her estate after death) then seeks to revoke the promise. Thesefacts have presented a problem for Anglo-American law because thetransaction, being gratuitous, is not enforceable as a contract. Property hasnot been legally transferred; neither is the promisor guilty of any tort.Nevertheless the courts of equity gave a remedy to the promisee.89 Thesecases could not be reconciled with orthodox contract doctrine respectingconsideration and have therefore been ignored or marginalised by manywriters on contract law. They have usually been described by commenta-tors as cases of proprietary estoppel but this phrase is scarcely explanatory.In some of the cases avoidance of unjust enrichment (though not always by

85 [1999] 3 SCR 108, 176 DLR (4th) 257.86 Ibid, at [24].87 Ibid, at [40].88 Ibid, at [25].89 Dillwyn v Llewelyn (1862) 4 De G F & J 517, 45 ER 1285 (CA); Ramsden v Dyson

(1866) LR 1 HL 129; Wilmott v Barber (1880) 15 Ch D 96; Plimmer v Wellington (1884) 9App Cas 699 (PC); Inwards v Baker [1965] 2 QB 29 (CA); Crabb v Arun DC [1976] Ch 179(CA); Greasley v Cooke [1980] 1 WLR 1306 (CA); Stiles v Tod Mountain Devt Ltd (1992) 64BCLR (2d) 366 (SC).

70 Stephen Waddams



that name) was evidently a crucial factor: in the leading case of Dillwyn vLlewelyn, for example, the plaintiff had expended the very large sum of£20,000 in improving land that was originally worth only £1,500.90 Oneof the considerations in the mind of a court faced with such facts has beenthe enrichment that would enure to the defendant if no measure ofenforcement were available, and one of the reasons for favouring propri-etary estoppel as a rule is that unjust enrichment is very apt to occur insuch circumstances, and so the rule tends to prevent unjust enrichment. Butunjust enrichment has not been present in every particular case,91 and theremedy has not normally been measured by enrichment. As Peter Birkswrote, ‘the doctrine … has a dimension to it which has nothing to do withrestitution/unjust enrichment’.92 This is true, but it does not follow thatconsiderations of unjust enrichment have been irrelevant. Many of thecases have had the effect of protecting reliance, but where the plaintiffbecomes effectively the owner of the land the measure of recovery is notrestricted to out of pocket loss. Non-contractual reliance has sometimesbeen protected by concepts of wrongdoing, but in these cases there is nowrongdoing in the ordinary sense. Though the phrase ‘equitable fraud’ hassometimes been employed, no actual proof of wrongdoing has beenrequired: the defendant acts fraudulently, in the eyes of equity, by failing todo what is just. As expressed in Willmott v Barber (1880), ‘the plaintiffmust prove that he [the defendant] has acted fraudulently, or that there hasbeen such an acquiescence on his part as would make it fraudulent for himnow to assert his legal rights’.93

The word ‘fraudulent’ in the last clause of this passage means ‘unjust’,and cannot be explained except in terms of concepts other than wrongdo-ing. The defendant must, by action or inaction, induce the plaintiff’sreliance, but no proof of intention to mislead or deceive is required.94 Theonly ‘fraud’ required to be proved is an unwillingness to do what equityconsiders just. A similar comment may be made in relation to the conceptof unconscionability.95 If equity protects the plaintiff’s reliance, it will be,

90 Dillwyn v Llewelyn (1862) 6 LT 878 (CA) 879.91 See J Beatson, Anson’s Law of Contract, 27th edn (Oxford, Oxford University Press,

1998) 119.92 P Birks, Introduction to the Law of Restitution (Oxford, Clarendon Press, 1985) 290

(emphasis added). R Goff and G Jones, The Law of Restitution (London, Sweet & Maxwell,1965) and P Maddaugh and J McCamus, The Law of Restitution (Toronto, Canada LawBook, 1990) included discussion of these cases in their books, but A Burrows, The Law ofRestitution (London, Butterworths, 1993) wrote flatly at 404 that these cases ‘have not beenrestitutionary’.

93 Willmott v Barber (1880) 15 Ch D 96, 106 (emphasis added); Gerrard v O’Reilly(1843) 3 D & War 414 (Ir Ch).

94 Willmott v Barber, ibid, at 105 (Fry J). This is a common usage in equity. See LSheridan, Fraud in Equity (London, Sir Isaac Pitman and Sons, 1957).

95 See M Spence, Protecting Reliance: the Emergent Doctrine of Equitable Estoppel(Oxford, Hart Publishing, 1999) 55–66; Giumelli v Giumelli (1999) 161 ALR 473 (HCA).




by that very fact, against conscience for the defendant to defeat it.‘Inequitable’, ‘fraudulent’, ‘unconscionable’ and ‘unconscientious’ havebeen, in this context, four ways of saying the same thing.96

Where a party to a contract indicates that strict rights will not beenforced, equity has in some cases prevented that party from resuming thestrict rights where it would be inequitable to do so. In Hughes vMetropolitan Railway a landlord who was entitled to demand repairs onsix months notice, gave the notice, but then indicated by conduct that therunning of the period of notice was suspended. The House of Lords heldthat the landlord could not enforce its strict rights without giving thetenant a reasonable opportunity of compliance. ‘Principle’ might bethought to permit the landlord to assert its strict rights, because there wasno consideration for the implied promise to suspend them, but the courtrelied on another proposition, also called a principle—indeed actuallycalled ‘the first principle’:

[I]t is the first principle upon which all Courts of Equity proceed, that if partieswho have entered into definite and distinct terms involving certain legalresults—certain penalties or legal forfeiture—afterwards by their own act or withtheir own consent enter upon a course of negotiation which has the effect ofleading one of the parties to suppose that the strict rights arising under thecontract will not be enforced, or will be kept in suspense, or held in abeyance,the person who otherwise might have enforced those rights will not be allowedto enforce them where it would be inequitable having regard to the dealingswhich have thus taken place between the parties.97

For ‘the first principle’, this is not very compendiously stated, and thereader might wonder how many other such ‘first principles’ there might be,whether they are a closed number, and whether they can all be definitivelyformulated.

In Central London Property Trust Ltd v High Trees House Ltd, alandlord had reduced the rent of an apartment building during wartime,and the question arose whether, after the war, the landlord could resume itsright to the full rent. Although the landlord made no claim for arrears, thejudge, Denning J, took the opportunity to say that arrears could not havebeen claimed. Denning J formulated the following proposition:

I prefer to apply the principle that a promise intended to be binding, intended tobe acted on and in fact acted on, is binding so far as its terms properly apply.98

This ‘principle’ had nowhere previously been formulated, and a few yearslater Denning himself (then a member of the Court of Appeal) was forced

96 See A Robertson, ‘Reasonable Reliance in Estoppel by Conduct’ (2000) University ofNew South Wales Law Review 87, 96–7.

97 (1877) 2 App Cas 439 (HL) 448.98 [1947] 1 KB 130, 136.

72 Stephen Waddams



to retreat substantially from it.99 The result of this advance and hastyretreat has been a very high degree of uncertainty about the scope of thealleged principle. Denning LJ used the word ‘principle’ five times in hisfairly short judgment,100 but it is not at all clear what principle the decisionapplied. Unanswered questions include whether a previous legal relation-ship between the parties is necessary, whether detrimental reliance isessential, whether estoppel can create new legal rights, whether thepromisee’s remedy is limited to protection of reliance, and whether strictrights can be resumed on reasonable notice. Denning tried again in 1982 toformulate a ‘general principle’ that would account for all instances ofestoppel:

The doctrine of estoppel is one of the most flexible and useful in the armoury ofthe law. But it has become overloaded with cases. . . . It has evolved during thelast 150 years in a sequence of separate developments: proprietary estoppel,estoppel by representation of fact, estoppel by acquiescence, and promissoryestoppel. At the same time it has been sought to be limited by a series of maxims:estoppel is only a rule of evidence, estoppel cannot give rise to a cause of action,estoppel cannot do away with the need for consideration, and so forth. All thesecan now be seen to merge into one general principle shorn of limitations. Whenthe parties to a transaction proceed on the basis of an underlying assumption—either of fact or of law—whether due to misrepresentation or mistake makes nodifference—on which they have conducted the dealings between them—neitherof them will be allowed to go back on that assumption when it would be unfairor unjust to allow him to do so. If one of them does seek to go back on it, thecourts will give the other such remedy as the equity of the case demands.101

But neither can this formulation be adjudged a success. In 1996 Millett LJsaid that the ‘attempt to demonstrate that all estoppels . . . are all governedby the same requirements has never won general acceptance’.102 In 2001Lord Goff said, of the statement of Lord Denning’s just quoted, ‘Thisbroad statement of the law is most appealing. I yield to nobody in myadmiration for Lord Denning; but it has to be said that his attempt in thispassage to identify a common criterion for the existence of various formsof estoppel . . . is characteristically bold.’103 Lord Goff then, having quoteda statement from a treatise to the effect that estoppel should not bepermitted to undermine the doctrine of consideration, added:

I myself suspect that this statement may be too categorical; but we cannot ignorethe fact that it embodies a fundamental principle of our law of contract. The

99 Combe v Combe [1951] 2 KB 215 (CA).100 Ibid, at 219–20.101 Amalgamated Investment and Property Co Ltd v Texas Commerce International Bank

Ltd [1982] QB 84 (CA) 122.102 First National Bank v Thomson [1996] Ch 231 (CA) 236.103 Johnson v Gore Wood & Co [2002] 2 AC 1 (HL) 39–40.




doctrine of consideration may not be very popular nowadays; but although itsprogeny, the doctrine of privity, has recently been abolished by statute, thedoctrine of consideration still exists as part of our law.104

Lord Goff then said, of estoppel, ‘in the end I am inclined to think that themany circumstances capable of giving rise to an estoppel cannot beaccommodated within a single formula, and that it is unconscionabilitywhich provides the link between them’.105 By this he meant that theunderlying reason for estoppel is that it would be unfair for a person,having made an assertion that induces another to act to her detriment, togo back on the assertion. This approach is capable, in effect, of enforcingpromises, as is illustrated by the Australian case of Waltons Stores(Interstate) Ltd v Maher.106 There an owner of land demolished a buildingand commenced construction of a new one to the specifications of aprospective tenant. No binding lease was ever effected, but the prospectivetenant was held to be estopped (in the view of the majority of the court)from retreating from an implied promise to complete the contract. Theeffect was that, although there was no contract, the prospective tenant wasbound by precisely those obligations that would have existed if there hadbeen a contract, a state of affairs that might not unreasonably besummarised by saying that, in effect, there was a contract.

‘Principle’ means beginning (principium), but it is evident that this is nota point in historical time: it is more akin to a conceptual or notional origin,root or source. From this it follows that principles are liable to varyaccording to the concepts or notions of the speaker or writer who uses theword. The search for principle in law is not a purely historical inquiry, butit has a historical dimension. The history of the law includes the history offailed attempts to formulate principles, of which we have seen severalinstances in the present context, and the questions of what concepts havebeen employed in the past, and how widely they have been held, arehistorical questions. Addison said that contract law was founded onprinciples that were immutable, eternal and universal, but the presentinquiry suggests that, in respect of consideration, it has not been possibleto formulate principles that have been stable even over a short period oftime in the history of one legal system. ‘Principle’ has, to say the least, beena very versatile tool, changing its shape to suit diverse purposes: in thehands of Viscount Simonds, principle prohibited reform; in the hands ofCrompton J, and of Lord Denning, principle demanded reform—but inentirely opposite directions. It does not follow that the concept of principlehas been useless or unimportant; on the contrary, the constant appeal by

104 Ibid, at 40.105 Ibid, at 41.106 (1988) 164 CLR 387 (HCA).

74 Stephen Waddams



courts and writers to principle suggests that the concept of principle,though not the formulation of any particular proposition, has beenperceived as essential to legal reasoning. A principle is a proposition thatlinks past, present and future: a proposition, to be accepted as a legalprinciple, must account satisfactorily for past decisions thought to havebeen rightly decided; it must resolve the current dispute in a mannerperceived to be satisfactory; and it must be capable of resolving in asatisfactory manner all future instances that can be envisaged. The appealto principle implies that propositions, even when it is known that they arebeing formulated and applied for the first time, can be traced to some sortof conceptual or notional beginning, and this implication has been animportant part of what has enabled an uncodified system to combineflexibility and the ability of the law to change, with the preservation ofcontinuity, stability, and coherence. Paradoxically, the concept of principlehas succeeded only by appearing to be what it is not. At each point of timeprinciples have appeared to be stable, as they must appear in order to becalled principles, but from a historical perspective it can be seen that theyhave been constantly susceptible to change.






4

Contractual Interpretation atCommon Law and Civil Law: An

Exercise in Comparative LegalRhetoric

CATHERINE VALCKE

THAT CIVILIANS AND common lawyers approach contractualinterpretation differently is well established.1 Also well established isthat, as with most issues of private law, the difference lies not so much

in the outcomes of the decided cases as in the means that are used to reachthese outcomes.2 That is to say, judges in the two systems reach outcomes thatare on the whole very similar, but they do so by deploying very different rulesand institutions. It is well-known, for example, that civil law judges oftenappeal to some general notion of good faith to arrive at conclusions whichcommon law judges instead reach by invoking implied terms.3

Now the particular reasons offered (or not offered) in support of thechoice of particular rules and institutions in each system are arguably farmore significant than the rules and institutions themselves. Because theweight and legitimacy given to legal decisions are largely determined by the

1 B Nicholas, The French Law of Contract (Oxford, Clarendon Press, 1992) 47ff; RDavid and D Pugsley, Les contrats en droit anglais (Paris, LGDJ, 1985) 251ff.

2 A Burrows and E Peel, ‘Overview’ in A Burrows and E Peel (eds), Contract Terms(Oxford, Oxford University Press, 2007) 3, 7–8; S Vogenauer, ‘Interpretation of Contracts:Concluding Comparative Observations’ in Burrows and Peel, ibid, at 123, 149–50. For onerecent Supreme Court of Canada decision where this difference might have nonethelessdetermined the outcome, see: Double N Earthmovers Ltd v City of Edmonton [2007] 275DLR (4th) 577 [Double N Earthmovers]. The case turned on the interpretation to be given totwo implied terms contained in a tendering contract. The five judges in the majority, three ofwhom were civilian (Lebel, Deschamps, and Fish JJ), favoured a narrow reading of the terms,which resulted in the tenderer not being found in breach; the four dissenting judges’ broaderreading would have resulted in a finding of breach. Had any one of the three civilian judgesendorsed the broader reading, judgment would have been rendered against the tenderer.

3 Nicholas, above n 1, at 49–50.



persuasiveness of the reasons offered in support of these decisions, legalactors intent on defending a particular decision can be expected to appealto the reasons that are considered most persuasive in their community. Andthe reasons considered most persuasive in any legal community naturallyare those that most directly tap into its core values, namely, its animatingconceptions of law, liberty, equality, the role of the State, and so on. Thereasons offered in support of legal decisions in any given community thuscan be taken to mirror that community’s fundamental legal values. What ismore, the bare fact that the legal actors appealing to these reasons perceivethem to be persuasive in their community is itself significant: beingthemselves members of that community, their perceptions as to thecommunity’s values to a certain extent are constitutive of those values.4

Verging on circularity, therefore, the reasons offered in support of legaldecisions are significant both because they reflect the community’s corelegal values and because the very fact that they are being heralded asreflective of these values makes them so.5 As the embodiment then of alegal system’s core values, the reasoning offered in support of particularchoices of rules and institutions is clearly useful for the purpose of comingto an internal understanding of that system taken as a whole. But it also isuseful for the purpose of understanding the rules and institutions them-selves. For instance, the reasons that common law judges offer forinterpreting contracts through implied terms say much about the funda-mental values of the common law system as a whole, as well as about theEnglish doctrine of implied terms. And then similarly the reasons given bycivilians in support of their judges resorting to good faith say much aboutthe fundamental values underlying the civil law, as well as about the natureof the civilian notion of good faith. For to ‘understand’ a legal rule is to beable to locate it within the larger edifice of the legal system to which itbelongs, to connect it with the other rules of that system; such a connectioncan only be established through the values reflected in the reasons given. Ameaningful comparative understanding of the various rules and institutionsdeployed in each different system for the purpose of interpreting contractstherefore will necessarily involve comparing these rules and institutions asthey relate to one another and to the other elements of their respectivesystems through the values embodied in the attendant reasoning.

The following account of contractual interpretation at common law andcivil law focuses accordingly on the reasoning used, that is, on the

4 This is not to say that the legal values of a community are nothing but a constructionof that community. It could well be that this layer of community-specific values issuperimposed upon or somewhat interwoven with another, deeper layer of universal valueswhich in contrast would by definition be common to all legal systems.

5 See generally N Luhmann, ‘L’unité du système juridique’ (1986) 31 Archives dephilosophie du droit 163.

78 Catherine Valcke



arguments that are offered, to justify contractual interpretation, and muchless upon the various rules and institutions of those systems. In that sense,this account qualifies as an exercise in comparative legal rhetoric. Its aim isthis: through the reconstruction of the respective conceptions of thecontractual enforcement process embodied in the arguments of eachsystem, the article seeks to connect the rules and institutions of each systemto one another and to contrast them between systems. These two tasks areaccomplished in tandem, not sequentially: as the reconstruction processunfolds and the rules on each side are being connected, the contrastbetween the systems becomes increasingly apparent. Part I describes andjustifies the parameters of this exercise. No comparative undertaking canproceed without first identifying an appropriate neutral common basis, ortertium comparationis, upon which to conduct the comparison.6 A com-mon language, conceptual territory, and set of criteria must be establishedwhich are sufficiently abstract to apply to the two terms under comparisonwithout distorting the identity of either, yet not so abstract as to bemeaningless. Part I aims accordingly to expose the neutrality of ourworking definition of ‘contractual interpretation,’ of the process by whichthe pool of relevant legal materials was delineated, and of the criteria usedto conduct the actual comparison. The comparison takes place in Parts IIand III, which are each devoted to surveying civil law and common lawmaterials on contractual interpretation. For reasons of convenience, thediscussion of Part II is limited to French and Quebec legal materials,7 andthat of Part III is limited to English and Canadian materials.

I. PRELIMINARIES

The legal materials targeted by the present study are those pertaining to‘contractual interpretation’ writ large, that is, to contractual interpretationunderstood as the task of determining the normative content of a contract.This task is distinguished from those of determining whether a contractexists in the first place, whether it was in fact breached, and if so, what isto be done to redress this breach. We will see that in both systems the term‘interpretation’ has been given a variety of more specific and at timesdivergent meanings. Precisely for that reason it is necessary to adopt, at

6 Since Radbruch (über die Methode der Rechtsvergleichung, MKSR II, 423, 1905/06),the notion of tertium comparationis is a staple of comparative law literature. See in particular:H Kötz, ‘Comparative Law in Germany Today’ [1999] Revue internationale de droit comparé753, 758ff.

7 A civil law survey that does not include German legal materials is obviouslyincomplete, but unlike Quebec materials, German materials are sufficiently different fromFrench materials to warrant a separate treatment. See eg, Vogenauer, above n 2, who runs theFrench and German analyses in parallel.

Contractual Interpretation 79



least provisionally, as broad a definition as possible, one which captures allthat is seen as pertaining to contractual interpretation in either system. For,as indicated, a study of legal rhetoric very much is an exercise in legalself-perception, an exercise in perception from the standpoint of the legalactors themselves.8

Among the materials on contractual interpretation so understood, spe-cial attention is given to those that feature legal actors choosing to supporttheir decisions with some particular arguments rather than others alsoreasonably available to them, and especially to materials that show legalactors choosing, from the pool of available arguments, those that seemleast plausible, all else being equal. For in the extreme case where only oneargument is reasonably available, the legal actor has no choice. The factthat that argument ends up being used only reflects the absence of analternative—it says nothing as to the actor’s assessment of the argument’srelative persuasiveness. In other words, persuasiveness being a matter ofdegree, judgments as to the persuasiveness of arguments are necessarilycomparative: they cannot but proceed from the observation of cases wherethe legal actor chooses one argument or set of arguments over at least oneavailable alternative. For example, in a contract case where the judgedecides in favour of the defendant on the ground that ‘the contract neverformed,’ the fact that the judge cites this reason is far more significant, forthe purpose of assessing this reason’s persuasiveness, if the facts are suchthat the alternative reason ‘the contract is unenforceable on ground ofunfairness’ is also reasonably available. For only then can it reasonably becontended that the judge might have cited the reason ‘the contract neverformed’ because the judge considered it more persuasive than the alter-native. Such a comparative judgment would simply not be possible werethe facts such that the reason ‘the contract never formed’ was the only onereasonably available to the judge. This example also underscores how thesignificance of any given choice of reasons, among a plurality, is inverselyrelated to the plausibility of the reason chosen. The likelihood that judgeschose ‘the contract never formed’ because they considered that reasonmore persuasive is greater where, all else being equal, that reason is, in thecircumstances, less plausible than the alternative. For it then is possible toconclude that that reason was chosen in spite of its being less plausiblethan the alternative, whereas one would normally expect judges to choose

8 The legal actors at play here are idealised legal actors in the same way that Dworkin’sHercules is an idealised judge (R Dworkin, Taking Rights Seriously (Cambridge, HarvardUniversity Press, 1977) 184ff): they are infinitely intelligent and knowledgeable of the legalmaterials that make up their respective systems, they take these materials seriously, as the soleembodiment of legal authority in their system, and they view themselves as charged with thetask of reconstructing these materials into a coherent conceptual entity. It is in that sense thatexercises in legal rhetoric differ from exercises in, say, legal anthropology or legal sociology,which in contrast aim to study actual legal actors.

80 Catherine Valcke



the most plausible among the available reasons. In sum, the less plausible achosen reason, the more significant its choice becomes, and the more likelyit is that that choice reflects a judgment about the relative persuasiveness ofthe various available reasons (and hence also the values peculiar to thelegal system). The survey of Parts II and III accordingly focuses, to theextent possible, upon materials that show legal actors choosing a particularargument or set of arguments over other available ones, with specificattention being paid to cases where the chosen arguments were lessplausible than their alternative.

The ‘legal actors’ under observation in Parts II and III mainly are thejudges, on the English side, and la doctrine (the scholars), on the Frenchside. While the decisions of all legal actors, not just those of judges orscholars, admittedly are relevant for the purpose of assessing what argu-ments are considered persuasive in the system,9 there are good reasons hereto focus on this particular set of legal actors. As a matter of convenience,English judges and French scholars remain the richest sources of easilyaccessible reasons in their respective systems. As a matter of principle,moreover, the tighter focus upon the legal actors considered formal orquasi-formal lawmakers in their respective systems is desirable for thepurpose of keeping the impartiality of the comparison in check: it ensuresthat the comparison is directed at dimensions of the two systems that canreasonably be considered counterparts of one another.10 And it indeed canreasonably be asserted that French scholars for present purposes are toFrench law what English judges are to English law. French judicialdecisions are notoriously short on reasons and therefore it has traditionallyfallen to the scholars to supply them.11

As for the neutral criteria with which to conduct the comparison, theyare to be found in the common functions discharged by contract law across

9 The arguments provided by legislators in support of their legislative decisions, byprivate parties in the context of contractual negotiations, by law students in support of theirexamination answers etc, can similarly be expected to be those considered most persuasive inthe community, and thus reflective of the community’s core values. See C Valcke, ‘“Precedent”and “Legal System” in Comparative Law: A Canadian Perspective’ in E Hondius (ed),Precedent and the Law (Brussels, Bruylant, 2007) 85.

10 Radical legal cultural relativists would deny the very possibility of establishing such‘counterparts’ between legal systems, but their argument, pressed to its logical conclusion,would also rule out the very possibility of any form of inter-system comparison, despite someof them having provided highly illuminating comparisons. See eg, P Legrand, Le droitcomparé (Paris, PUF, 1999).

11 This explains the great authority ascribed to scholars at French law—certainly ascompared to scholars at English law, but also as compared to judges at French law—similar tothat enjoyed by the great jurists of Roman law. See generally RC Van Caenegem, Judges,Legislators and Professors—Chapters in European Legal History (Cambridge, CambridgeUniversity Press, 1987) 67–111.




legal systems.12 If only intuitively, it indeed is possible to think of contractlaw in all systems as serving at least two different and somewhatcontradictory functions. The first is to ‘enforce’ contracts: strictly speaking,that is to bring about whatever it is that private parties want to be broughtabout. This first function can be described as ‘consecrating’, in so far as itentails doing no more than supplying the parties’ original intention13 withlegal force, to elevate that intention from the realm of fact into the realm oflaw. The parties’ own original conception of their respective obligationshere is adopted (and consecrated), at least aspirationally, as is; no attemptis made to impose on the parties anything that they did not originallyintend. The second function, which I call ‘disciplining,’ instead involvesrunning counter to the parties’ original intention. Rather than giving (legal)voice to the parties, the aim here is to dictate what the parties are allowedto intend, to substitute an acceptable legal intention for the parties’ own. Inthat sense, the disciplining function is prescriptive, unlike the consecratingfunction, which aspires to remaining normatively neutral. The consecratingand disciplining functions are conceptually contradictory in so far as thefirst involves moving along with the parties, whereas the second entailspressing back against them.14

In practice, be it civil law or common law, the consecrating anddisciplining functions of contract law cannot be so neatly distinguished,because both these functions inevitably are simultaneously at play in theenforcement of actual contracts. The contractual intention being enforcedindeed always is the parties’ original intention as it has been processed bythe court: the parties provide the raw data which the court then proceedsto re-formulate, re-cast, transform into something deserving of the law’s

12 The focus upon the common functions discharged by the different legal rules andinstitutions in the different legal systems as the starting point for comparative analysis isknown in comparative law literature as ‘functionalism’. Although functionalism as a theory ofcomparative law has been much criticised (L-J Constantinesco, Traité de droit comparé (Paris,LGDJ, 1983) vol III, 63–71) its value as one of several methodological tools available to thecomparatist is widely acknowledged: J Reitz, ‘How to do Comparative Law’ (1998) 46American Journal of Comparative Law 617, 620–23; G Samuel, ‘Epistemology and Com-parative Law: Contributions from the Sciences and Social Sciences’ in M Van Hoecke (ed),Epistemology and Methodology of Comparative Law (Oxford, Hart Publishing, 2004) 35,38ff.

13 It here is assumed that one such common intention exists. This assumption will berelaxed below.

14 Although the consecrating/disciplining distinction obviously parallels the continentaldistinction between ‘subjective’ and ‘objective’ right (Droit/Rechts), these terms here areavoided both because they resonate differently at French and English law and because thecontinental distinction carries much baggage that here is superfluous and thus potentiallymore confusing than useful. The present use of the consecrating/disciplining distinction astertium comparationis is justified, despite this distinction being connected with the continen-tal one, in so far as, in this new incarnation, and with the caveat just mentioned, it is arguablysufficiently abstract to be usefully applicable to both systems without distorting either.

82 Catherine Valcke



imprimatur. The final product therefore always is the result of a combina-tion of consecrating and disciplining.15 As indicated, however, it is not thefinal product of the contractual enforcement process in the two systemsthat interests us here, but rather how that process is viewed by the legalactors in these systems. And the views of the legal actors indeed appear todiverge in this regard, as I propose to show next. The view that emergesfrom the civil law materials is that of a process in which the consecratingand disciplining functions are neatly delineated from one another, and thedisciplining function is by far the more prominent (Part II). The view thatemerges from the common law materials in contrast is that of a process inwhich the two functions are merged, or more specifically, in which thedisciplining function has been merged into the consecrating function in anapparent attempt to downplay the former’s relative significance (Part III).

II. (FRENCH) CIVIL LAW MATERIALS ON CONTRACTUALINTERPRETATION

The prominence of the disciplining function in the French conception ofcontract law is obvious from the outset, from the way that the contractualenforcement process is structured. As one must have determined that acontract exists before undertaking to determine its content, contractadjudication at French law, just like at English law (and any otherminimally logically-inclined law), proceeds in that order. Also like atEnglish law, moreover, contract formation is considered an issue to beobjectively determined by the law makers, that is, independently from theparties’ own views on the matter.16 That is to say, French and English legalactors alike consider that it falls to the law makers, not the parties, to fixthe conditions of existence of a contract and to determine whether theseconditions are satisfied in any given case.17 Inasmuch as these conditions,and the determination as to their satisfaction, can accordingly be said to be‘imposed’ on the parties, it can be said that the issue of contract formationin both systems is treated as a ‘disciplining’ rather than a ‘consecrating’

15 Even in cases where the court apparently ends up enforcing the parties’ intention as is,the disciplining function is present since the court then implicitly signifies its normativeapproval of that intention.

16 The fact that consent—perhaps the most fundamental of the constitutive elements ofthe contract—itself is, as we will see, largely determined subjectively does not change the factthat the qualification of consent as such an element itself is the result of an objectivedetermination.

17 In so far as the ‘gentlemen’s agreements’ cases of English law are seen as cases in whichthe court refuses enforcement because the parties declared that their agreement ought not beconsidered a binding contract, these cases would constitute an exception. See, eg, Rose andFrank Company v JR Crompton & Brothers Ltd [1923] 2 KB 261 (CA). But other readings ofthese cases are possible which justify non-enforcement on the basis that one or several of the(objectively determined) elements of a contract in fact are missing in these cases.




issue under the above definition. The two systems, however, differ in that,at French law (unlike at English law) a determination of the issue ofwhether a given contract was formed also determines in large part thedistinct and subsequent issue of what that contract contains, with theresult that the latter issue is, to a large extent, also determined objectively.

The existence of a convention at French law indeed is determined by thepresence of one more elements than those familiar to the English lawyer.French law requires, in addition to an échange de consentements (offer andacceptance), the capacité de contracter (capacity to contract) of the parties,and a cause (roughly, consideration18), that there be ‘a determinate objet(object) forming the subject matter of the agreement’.19 In any contractdispute, therefore, the ‘subject matter of the agreement’—also known asthe ‘juridical operation envisaged by the parties’20—is to be objectivelyestablished at the outset, as one of the conditions for the existence of theconvention. But this identification of the objet for the purpose of establish-ing the existence of the convention also serves to determine much of thecontent of that convention, for it allows for both an early check on itslicéité (legality) and its initial classification. Unlike at English law, wherethe control of the contract’s legality cannot but intervene after the existenceof the contract has been established,21 the objet as a constitutive element ofa convention at French law allows for one such control to occur at theinitial stage of formation: contracts with illicit subject matters are deemednon-existent from the outset.22 In addition, the initial identification of theobjet allows for the classification of the convention under one of many

18 It is well known that the French cause differs substantially from the English notion ofconsideration. Whereas the notion of consideration entails the mutuality of a bargain, theFrench cause has been interpreted broadly, so as to be deemed present even in gifttransactions, in the form of ‘donative intent’. French jurists have debated the issue adnauseam, some of whom (the ‘anti-causalists’) suggesting that such a broad understanding ofcause is tantamount to none at all, and that the whole notion therefore might as well beabandoned. See eg, G Ripert and J Boulanger, Traité de droit civil (Paris, LGDJ, 1956–59) vol2, § 287.

19 ‘Un objet certain qui forme la matière de l’engagement.’ Art 1108 CC. (Mytranslation—all English translations of French texts are mine unless otherwise indicated.) Art1371 of the CCQ similarly provides that ‘[i]t is of the essence of an obligation that there bepersons between whom it exists, a prestation which forms its object, and, in the case of anobligation arising out of a juridical act, a cause which justifies its existence’. (The CCQ,unlike the Code Napoléon (CN), has an official English version.)

20 Art 1412 CCQ; F Terré, P Simler and Y Lequette, Droit civil—Les Obligations, 5thedn (Paris, Dalloz, 1993) §§ 257, 287.

21 Where the court concludes from its examination of the content of the contract that thiscontract is illegal or against public policy, the contract is declared to be retroactively void abinitio. See Archbolds (Freightage) Ltd v Spanglett Ltd [1961] 1 QB 374 (CA).

22 Art 1413 CCQ provides that ‘[a] contract whose object is prohibited by law orcontrary to public order is null’. At French law, in contrast, the licéité requirement of the objetwas articulated by la doctrine (see, eg, Terré et al, above n 20, at § 304); it is not explicit inthe Code.

84 Catherine Valcke



‘types’ for which the Code provides a list of pre-packaged terms,23 some ofwhich are waivable by the parties (dispositions supplétives) while othersare not (dispositions impératives).24 Even contracts initially classified as‘innominate’—contracts that fail to qualify under any of the availablenominate contracts—are typically presumptively governed by the rulesapplicable to the nominate contract(s) that they most resemble.25 Onlyonce this initial stage of classification is completed does the court launchinto the stage of interprétation proper, that is, turns to consider the termsof the particular contract at hand to determine the extent to which, if atall, the parties might have intended to depart from some of the dispositionssupplétives applicable to their contract type.26 Much of the normativecontent of the contract is thus already (objectively) determined by the timethe court reaches the stage of interprétation.

To be sure, some room at that stage is made for the parties’ ownintentions, inasmuch as that stage begins, as suggested, with an inquiryinto the extent to which the content which the parties intended to assign totheir contract might differ from the objective content presumptively fixedby the Code. The intention of the parties there under examination is theirsubjective—or ‘actual’, ‘psychological’, ‘private’ or ‘internal’—intention,27

23 The section of the French Civil Code describing the ‘Essential Conditions of Validity’generally applicable to all conventions is accordingly supplemented by ‘the rules particular tocertain contracts established under the titles relating to each one of them’ (Art 1107 CC).Thereafter appears the list of 15 said titles, covering Arts 1387–2203 CC (Title IV—Marriage;Title VI—Sale; Title VII—Exchange; Title VIII—Lease; Title IX—Incorporation; TitleX—Loan; Title XI—Deposit; Title XIII—Mandate; Title XIV—Suretyship; Title XVI—Settlement etc). The CCQ similarly provides for 18 nominate contracts. See TitleII—Nominate Contracts of Book Five—Of Obligations, covering Arts 1708–2643 CCQ.

24 For example, Art 1388 CC, in the Title IV—Marriage, provides that, in a marriagecontract, ‘the spouses cannot derogate either from the rights and obligations accruing to themas a result of the marriage or from the rules governing parental authority, legal administra-tion, and tutorship’. The CCQ is more explicit, stating more generally that ‘[i]n the exercise ofcivil rights, derogations may be made from those rules of this Code which supplementintention, but not from those of public order’ Art 9 CCQ.

25 Nicholas, above n 1, at 49.26 As Nicholas aptly explains: ‘The French starting-point is that the incidents of a

contract are fixed by law, subject to the parties’ power to vary them. … French law beganwith the Roman system of typical contracts and superimposed on it the unitary consensualprinciple that any agreement is a contract.’ Nicholas, above n 1, at 49. It has been suggestedthat the Roman system of typical contracts was similarly reproduced in the English writsystem: W Buckland and A McNair, Roman Law and Common Law (Cambridge, CambridgeUniversity Press, 1936) 204ff. Although this is to a certain extent the case, the Englishsimilarity with the Roman system is much less significant for our purposes in that it, unlikethe French, was more the result of a contingent combination of institutional circumstancesthan that of some form of conscious juridical deliberation. In addition, the establishment ofthe cause of action hardly determined the normative content of the parties’ transaction to thesame extent that it does at French law (ie, with a whole package of clearly defined terms).

27 J Flour and J-L Aubert, Les obligations—1. L’acte juridique, 5th edn (Paris, ArmandColin, 1991) § 189; Terré et al, above n 20, at § 228; B Starck, H Roland and L Boyer, Droitcivil: les obligations, 6th edn (Paris, Litec, 1999) 192; H Mazeaud, J Mazeaud and F Chabas,




as comparatists have often outlined.28 That is to say, any evidence as towhat the parties actually and possibly individually intended their contractto contain, explicitly or implicitly, would in principle be consideredrelevant, even such evidence as might not have been formally shared withthe other party at the time of forming the contract. French jurists have longconsidered that the subjective conception of contractual intention islogically entailed by the moral ideal of the ‘autonomy of the will’, whichtraces the source of contractual obligation to the will of the individual asself-governing agent.29 At any rate, that is the justification which theyprovide in support of the code dispositions that they see as embodying thisideal, in particular, the foundational article 1134 CC: ‘legally formedconventions are to be considered as law by the parties who made them’.30

As this justification for inquiring into the parties’ subjective intention tapsstraight into the heart of the consecrating function as defined above,French contractual interprétation thus far can be seen as consisting of apocket of consecrating in the otherwise unambiguously disciplining largerframework of contract law.

It has been observed that French jurists may be mistaken in thinking thatthe subjective conception of intention logically follows from the ideal ofthe autonomy of the will,31 and that French law in practice is not nearly as

Leçons de droit civil, 8th edn (Paris, Montchrestien, 1986) 161. For Quebec authors to thesame effect: J Pineau, D Burman and S Gaudet, Droit des obligations (Montreal: Thémis,2006) esp §§ 1696–99.

28 K Zweigert and H Kötz, Introduction to Comparative Law (Oxford, Clarendon Press,1987) vol II, 83–94; D Harris and D Tallon, ‘General Introduction’ in Harris and Tallon (eds),Contract Law Today—Anglo-French Comparisons (Oxford, Clarendon Press, 1989) 1–5; JCartwright, ‘Defects of Consent and Security of Contract: French and English Law Com-pared’ in P Birks and A Pretto (eds), Themes in Comparative Law in Honour of BernardRudden (Oxford, Oxford University Press, 2002) 156–7; M-A Frison-Roche, ‘Remarques surla distinction de la volonté et du consentement en droit des contrats’ [1995] Revuetrimestrielle de droit civil 573; A Rieg, Le rôle de la volonté dans l’acte juridique en droit civilfrançais et allemand (Paris, LDGJ, 1961); Nicholas, above n 1, at 47–9.

29 ‘French law consecrates the omnipotence of the actual will of the author of adeclaration of will. This is just the logical consequence of the principle of the autonomy of thewill.’ J Chabas, De la déclaration de volonté en droit civil français (Paris, Sirey, 1931) 81–2.See generally E Gounot, Le principe de l’autonomie de la volonté en droit privé, étude critiquede l’individualisme juridique (Paris, A Rousseau, 1912); V Ranouil, L’autonomie de lavolonté, naissance et évolution d’un concept (Paris, PUF, 1980); G Rouhette, ‘The ObligatoryForce of Contract in French Law’ in Harris and Tallon, ibid, at 38–40. For similar remarks inQuebec doctrine: Pineau et al, above n 27, at §§ 35, 154–6, 223–8.

30 ‘Art. 1134. Les conventions légalement formées tiennent lieu de loi à ceux qui les ontfaites …’ The equivalent CCQ provisions are Art 1378 (‘A contract is an agreement of willsby which one or several persons obligate themselves to one or several other persons toperform a prestation.’) and Art 1386 (‘The exchange of consents is accomplished by theexpress or tacit manifestation of the will of a person to accept an offer to contract made tohim by another person’).

31 C Valcke, ‘Objectivisme et consensualisme dans le droit français de l’erreur dans lesconventions’ (2005) 2 Revue de la Recherche Juridique 661.

86 Catherine Valcke



subjective as French jurists portray it to be.32 If anything, though, thisobservation would serve to reinforce the portrayal of French contractualinterpretation just given. As our objective is merely to identify thedistinctively French rhetoric that emerges from the materials understudy—to come to see these materials in the way that French legal actorsthemselves see them—the fact that their conception of contractual interpre-tation might somehow be mistaken or unsupported in practice is notimmediately relevant. Inasmuch as the justification that French legal actorsprovide in support of that conception shows that they see it as fostering theconsecrating purpose of contract law, it is possible to say that theconsecrating dimension of contract law is important to the French.33 Asexplained in Part I, moreover, the fact that French legal practice might situneasily with the subjective conception only serves to confirm the impor-tance that they attach to that conception, for it is an example of legalactors insisting on a particular reasoning despite the availability of apossibly more plausible alternative (the objective conception). Preciselybecause the subjective conception of contractual interpretation may bequestionable on logical or practical grounds, it is possible to say of Frenchlaw that it ‘clings to the ideal of autonomy of the will’, and that it is‘therefore ‘subjective’ [only] in terms of ideology and rhetoric’.34

Even such a small pocket of consecrating might be more than one shouldcare to admit to, however. For that pocket is not just small: its veryexistence, as well as the modalities for establishing this existence, isultimately conditioned by the discipline of the larger framework. Like alllegal systems, French law contains rules of evidence that constrain themeans by which the parties’ (subjective) intention can be established incourt, many of which much resemble various strands of the parol evidencerule of English law.35 If only through the imposition of evidentiary

32 M Tancelin, Des obligations (Montreal, Wilson and Lafleur, 1988) § 215.33 The possibility of mistake here does not undermine the idealised conception of legal

actors at play. See above n 8. Even if logically mistaken, the subjective conception ofcontractual intention is in fact supported by French legal materials, that is, it is a mistake inlogic only, not in law. As such, this mistake would be one that even an idealised legal actorcould make.

34 Vogenauer, above n 2, at 127 (emphasis added). The fact that the French might ‘cling’to this ideal in the face of contradicting legal practice moreover is revealing as to the relativelylower significance which they attach, more generally, to legal practice as compared to abstractmoral discourse.

35 Much like s 4 of the English Statute of Frauds and the English parol evidence rule moregenerally, Art 1341 (1) CC provides that contracts involving sums in excess of €1500 must beproven through the production of a signed or notarised document, and that such proof maynot be contradicted through testimonial evidence. Some of the many exceptions to the rule ofArt 1341 (1) CC are listed in the Code itself (Arts 1341(2)–1348 CC); others were developedby la jurisprudence (Civ 31 May 1948 S 1949 1.127) and la doctrine (Terré et al, above n 20,at § 153; J Ghestin and G Goubeaux, Traité de droit civil: introduction générale, 4th edn(Paris, LGDJ, 1994) §§ 663–4). Similar rule and exceptions are found in the CC Q at Arts2860–68.




constraints, the disciplining framework retains a hold on the consecratingpocket. More significantly, however, the consecrating justification thatFrench jurists offer in support of their reverence for party intention ispresented as a moral justification—as one that is extra-legal.36 Of course,from the moment that this justification is articulated by these jurists in thecourse of legal arguments, it is no longer strictly moral, particularly giventhat the jurists are themselves considered a quasi-formal source of law.37

Strikingly, however, the jurists’ appeal to the moral ideal of the autonomyof the will is almost always made concurrently with an appeal to article1134 CC,38 which suggests that they feel justified in appealing to that idealonly in so far as it can be shown to have been embodied into objectivelaw.39 That is to say, French jurists view the parties’ intention as beingderivatively, not inherently, legally normative—as being legally normativeonly in so far as it qualifies as one of those facts to which the law hasdecided to extend its normativity. This is confirmed by the fact that thedeciphering of party intention at French law is in principle considered a(mere) issue of fact, unlike the earlier stage of classification of the contract,which is considered an issue of law.40 Therefore, our small pocket ofconsecrating nonetheless remains thoroughly conditioned by the largerdisciplining framework.

Thus far our discussion has focused upon the dominance of thedisciplining function in the French conception of contract law. But thatdiscussion also exposes the consecrating and disciplining functions asstrictly delineated from one another. For subjective contractual intention—what the parties actually intended—is a conceptually clear and distinct

36 See, eg, Flour and Aubert, above n 27, at § 94; Terré et al, above n 20, at §§ 19–20;Pineau et al, above n 27, at § 154.

37 Above n 11.38 See, eg, Flour and Aubert, above n 27, at § 95; Terré et al, above n 20, § 5; Pineau et

al, above n 27, at §§ 155, 223.39 Flour and Aubert’s following comment (ibid) on Art 1134 (1) CC is particularly

eloquent in this respect: ‘There is no need to rehash here the excessive character of this‘equalisation’ of the contract with the law: in reality, the latter is superior to the former, sincethe former’s validity always is subordinated [to it].’

40 Ch réunies 2 February 1808 S 1808 1.480; D Lluelles and B Moore, Droit desobligations (Montreal, Thémis, 2006) §§ 1726ff; Tancelin, above n 32, at § 228; Terré et al,above n 20, at §§ 459ff. This entails that the lower courts’ interpretive rulings on contractualinterpretation in principle are not reviewable by higher courts, including the French supremecourt in matters of private law, the Cour de cassation. Institutional concerns that lower courtswere abusing this ‘sovereign power’ (pouvoir souverain) of theirs however led to theinstitution of the recours en dénaturation, which allows for higher court review in cases wherethe lower court’s interpretation of ‘clear and precise terms’ is so outrageous as to amount to a‘distortion’ (dénaturation) of the contract. Civ 15 April 1872 S 1872 1.232. Such recoursremain highly exceptional, however: J Voulet, ‘Le grief en dénaturation devant la Cour decassation,’ Jurisclasseur Périodique 1971.1.2410, [1]; M-H Maleville, Pratique del’interprétation des contrats: étude jurisprudentielle (Rouen, Publications de l’Université deRouen, 1991) §§ 149ff.

88 Catherine Valcke



notion, certainly tighter than its objective counterpart, the somewhatloose41 notion of what the parties might, or ought to, or could reasonablyhave intended. This is the case whether one considers explicit or implicitsubjective intention, in spite of the obvious practical difficulties attendingan investigation of implicit subjective intention. While subjective intention,especially in its implicit incarnation, indeed may at times be difficult toaccess, its determination remains free of the kind of substantive indetermi-nacy that besets an inquiry into objective intention. No question arises asto what is to count as ‘subjective intention’. Such intention, being a matterof fact to be determined through empirical investigation, all and only thatwhich was in fact intended by the parties is to count as ‘subjectiveintention’ for the purpose of legal consecration. The strict delimitation ofthe pocket of consecrating from the surrounding disciplining framework ishence made possible by the conceptual tightness of the notion upon whichit is built, namely, party intention as real, empirically determined intention.

That the French regard the pocket of consecrating as strictly delimited asbeing borne out by the fact that, as indicated, the determination of partyintention at French law is formally earmarked an issue of fact, whereas theother determinations making up the contractual enforcement process areearmarked issues of law. This also is evident from the rules governing themeans by which party intention is to be accessed and established in courtproceedings, the package of which is, as suggested, functionally roughlyequivalent to the parol evidence rule of English law. The French andEnglish packages indeed differ in their respective organisations. Unlike theEnglish package, which as a whole straddles evidentiary and substantivelaw, the French package neatly divides into its substantive and evidentiarycomponents. Rules pertaining to the classic canons of interpretation, mostof which were designed to guide the interpreter in accessing party intention(the contra proferentem rule, the rule ut res magis valeat quam pereat andso on), are considered rules of substantive law and accordingly found inthe section of the Code entitled ‘Of the Interpretation of Conventions’located in one of the chapters devoted to substantive contract law.42 In

41 That one notion is described as ‘tight’ and the other as ‘loose’ should not be read as asuggestion that the one is somehow ‘better’ or ‘superior’ or ‘more desirable’ than the other.Conceptual looseness may be indicative of greater complexity and richness, and it may entailgreater flexibility, which is generally considered a virtue from a legal perspective. Conversely,conceptual ‘tightness’ may be symptomatic of shallowness, bareness or rigidity. Simplisticnotions are conceptually tight, yet are no less defective for that.

42 Arts 1156–64 CC, contained in Section V of Chapter III of Title III (De l’interprétationdes conventions). The same rules are found in Section IV (‘Interpretation of Contracts’) ofChapter II (‘Contracts’) of Book Five (‘Obligations’) of the CCQ. See, eg, Art 1427 (‘Eachclause of a contract is interpreted in light of the others so that each is given the meaningderived from the contract as a whole.’); Art 1428 (‘A clause is given a meaning that gives itsome effect rather than one that gives it no effect.’); Art 1430 (‘A clause intended to eliminatedoubt as to the application of the contract to a specific situation does not restrict the scope of




contrast, rules pertaining to evidence per se, form a chapter of their ownentitled ‘Of the Proof of Obligations, and of their Payment’.43 Frenchdoctrinal treatments of these issues moreover exhibit the same segregatingtendency: it is not uncommon to find purportedly general discussions ofcontractual interpretation that are completely silent on evidentiary issues,44

and where those issues are discussed, it is typically in a section separatefrom that devoted to the substantive rules.45 Conversely, no French treatiseon the law of evidence that could be found even alludes to the substantiverules.46

The content of the substantive rules further confirms the conceptualtightness of subjective intention and resulting strict delimitation of theconsecrating pocket from the surrounding disciplining framework. Unlikethe evidentiary rules, which, as explained, cannot but be disciplining innature, the substantive rules contained in the interpretation section of theCode are not intended to modify, add to, or subtract from the parties’actual intention in any way. Consistently with the empirical conception ofintention as made up of all and only that which the parties actuallyintended, those ‘rules’ are in fact mere suggestions, interpretive guide-lines,47 codified for no other purpose than to assist the courts in theirdeciphering task.48 In the same vein, French law has never placed a priorilimitations on the admissibility of materials deemed relevant for thatpurpose: as alluded to above, all materials likely to help shed light on theparties’ actual intention in principle are admissible, including materialspertaining to such contextual factors as pre-contractual, collateral, or

a contract otherwise expressed in general terms.’); Art 1432 (‘In case of doubt, a contract isinterpreted in favour of the person who contracted the obligation and against the person whostipulated it.’).

43 Arts 1315–69 CC, contained in Chapter VI of Title III (De la preuve des obligations etde celle du paiement). The CCQ is even more eloquent in this regard, as it segregates thearticles relating to evidentiary issues (Arts 2803–74 CCQ) into a book of their own (‘BookSeven—Evidence’), altogether distinct from the general book on obligations (‘Book Five—Obligations’).

44 See, eg, Maleville, above n 40; Pineau et al, above n 27, at § 224 (who devote four linesto evidentiary issues).

45 See, eg, Terré et al, above n 20; Ghestin and Goubeaux, above n 35.46 See, eg, C Perelman and P Foriers, La preuve en droit (Brussels, Bruylant, 1981); CE

Dorion, De l’admissibilité de la preuve par témoins en droit civil (Montreal, Whiteford &Théoret, 1894).

47 Vogenauer explains that the German codifiers, unlike the French, deliberately refrainedfrom including those rules in their code precisely because they considered them to be ‘nothingbut rules of logic’. Vogenauer, above n 2, at 130–31.

48 ‘pieces of advice given to the judges, in matters of contractual interpretation, morethan strict mandatory rules from which they may never depart, not even in the mostcompelling of circumstances’ Req 18 March 1807 S 1807 1.361 (C). See generally Maleville,above n 40, at §§ 260ff.

90 Catherine Valcke



subsequent statements or actions by the parties,49 even such materials asmay pertain to one party’s intention undisclosed to the other.50 Thisexplains why the debate over literal versus contextual interpretation has, inFrance unlike in England, focused almost exclusively upon statutory, notcontractual, interpretation:51 as article 1156 CC bluntly confirms,52 literal-ism simply is antithetical to the subjective conception of contractualinterpretation. This also explains why French law does not have anythingresembling the English doctrine of rectification: there is no need for somecorrective measure to bring contractual interpretation in line with partyintention in a system where these are one and the same.53 A small pocketof consecrating contained in and conditioned by a larger discipliningframework, then, yet one that can nonetheless be clearly delineated fromthat framework.

The strict delineation between the consecrating and disciplining func-tions, and the dominance of the latter, are further reiterated in the secondand final phase of the interpretation stage. The reader will remember that,following the operation of classification of the contract, at the stage ofcontract formation, the court moves to the interpretation stage proper, thefirst phase of which, as just explained, consists in the court deciphering theparties’ explicit and implicit intention. The second phase involves the courtdetermining what to do with that intention. In particular, the court thendetermines which parts of that intention ought to be retained, which oughtto be discarded, and how that intention is to be supplemented, all with aview to processing that intention into something worthy of legal authority.As indicated, of the many provisions presumptively assigned to theircontract type under the Code, the parties are allowed to depart only fromthose deemed dispositions supplétives, but not from those deemed disposi-tions impératives.54 The court accordingly includes party departures from

49 ‘[One must] consult at the same time the terms of the act, the circumstances thatpreceded it, those that followed it, …’ Req 9 May 1877 DP 1878 1.30, cited in Maleville,above n 40, at § 403. See generally, §§ 403–42. The same is true under Quebec law: Lluellesand Moore, above n 40, at §§ 1599ff.

50 Above text accompanying nn 27–9.51 P-A Côté, Interprétation des lois (Montreal, Yvon Blais, 1982) 18ff. To the extent that

the teachings of the exégètes did transpire from statutory to contractual interpretation, theywould be reflected in the exceptional remedy of the recours en dénaturation. See Vogenauer,above n 2, at 131–2.

52 ‘One must in contracts seek to ascertain what was the common intention of thecontracting parties, rather than stop at the terms’ literal meaning.’ The equivalent CCQ articlereads: ‘The common intention of the parties rather than adherence to the literal meaning ofthe words shall be sought in interpreting a contract.’ Art 1425 CCQ.

53 See, eg, Ghestin’s discussion of the famous German case involving parties who meantto buy and sell whale meat but wrote ‘shark meat’ in the contract: J Ghestin, Traité de droitcivil: la formation du contrat, 4th edn (Paris, LGDJ, 2000) § 495.

54 Above n 24.




the former, but not party departures from the latter. The court alsodisregards clauses otherwise deemed ‘abusive’,55 which the Civil Code ofQuebec (CCQ) defines as follows:

An abusive clause is a clause which is excessively and unreasonably detrimentalto the consumer or the adhering party and is therefore not in good faith; inparticular, a clause which so departs from the fundamental obligations arisingfrom the rules normally governing the contract that it changes the nature of thecontract is an abusive clause.56

At French law like at Quebec law, moreover, the notion of an ‘abusive’clause is semantically—thus also juridically57—related to the notion of‘abuse of rights’, the disciplining flavour of which is such as to startle anyEnglish lawyer. The latter notion indeed operates as a license for the courtsto find that particular actions by the parties, although technically withinthe ambit of their rights, nonetheless are to be denounced as illegitimateexercises of these rights.58

Following article 1135 CC, moreover, the court also reads into thecontract ‘all that ought to follow from the nature of the obligation underequity, usage, and the law’.59 The main such obligation ‘following fromequity’ is the famous obligation to execute contractual obligations in goodfaith, codified at article 1134(3) CC60—‘a powerful filter’61 for onerous

55 Civ 1ère 6 déc 1989 D 1990.289 note J Ghestin, RTD civ 1990.277 obs J Mestre. Thelegal basis for that power is still contested: some judicial decisions invoke Art 1134 CC, whileothers refer to Art 35 of the Loi du 10 janvier 1978.

56 Art 1437 CCQ.57 The semantic-juridical connection flows from the deliberateness of the codification

process.58 See, eg, National Bank of Canada v Soucisse [1981] 2 SCR 339 (failure of bank to

inform heirs of revocability of suretyship provided by deceased); Houle v Canadian NationalBank [1990] 3 SCR 122, 74 DLR (4th) 577 (bank recalling loan and realising on its guaranteewithin hours of announcing its intention to do so). The French notion of abuse of rightssomewhat resembles, yet is not nearly as strictly framed as, the estoppel of English law. Seegenerally A Mayrand, ‘Abuse of Rights in France and Quebec’ (1973–74) 34 Louisiana LawReview 993.

59 The equivalent CCQ article reads: ‘A contract validly formed binds the parties whohave entered into it not only as to what they have expressed in it but also as to what isincident to it according to its nature and in conformity with usage, equity or law.’ (Art 1434CCQ.)

60 The equivalent CCQ article reads: ‘The parties shall conduct themselves in good faithboth at the time the obligation is created and at the time it is performed or extinguished.’ (Art1375 CCQ.)

61 P Delebecque and D Mazeaud, ‘Les clauses de responsabilité: clauses de non-responsabilité, clauses limitatives de réparation, clauses pénales’ in M Fontaine and G Viney(eds), Les Sanctions de l’inexécution des obligations contractuelles (Paris, LGDJ, 2001) 361,372.

92 Catherine Valcke



contractual clauses. As articles 1134 and 1135 are located in the Disposi-tions générales of Chapter III titled ‘Of the effect of obligations’, they aregenerally applicable to all contracts.62

The disciplining dominance at play in this second phase of the interpre-tive stage is obvious enough, in so far as this is the moment at which thelaw’s transformative power is applied to the parties’ intention. Theintention that ends up being legally consecrated is the parties’ originalintention as filtered, supplemented, or at the very least endorsed by thecourt. Of our original pocket of consecrating indeed not much is left.Many French scholars accordingly have described this second phase insuch terms as ‘[the stage at which] the common will of the parties is seizedas a pretext for justifying holding the parties to the rules of equity’.63

Further still, in many cases the justifications offered for such judicialinterventions do not even allude to the common will of the parties, whetheras pretext or otherwise.64 More often than not, those justificationsexplicitly appeal to such unambiguously disciplining notions as ‘the court’sdesire to protect the weaker party’.65 As some prominent observers haveconcluded, ‘[i]t therefore could not be better established that these [con-tractual] obligations draw their source, not from some professed will of theparties, but from the law’.66

This second phase also underscores the French insistence on keeping theconsecrating and disciplining functions strictly delineated. However littlemay be left after the second (disciplining) phase of the parties’ originalintention identified in the first (consecrating) phase, French jurists insist onkeeping these phases separate, not just sequentially, but also semantically:

62 The obligation of good faith contained at Art 1375 CCQ is even more general, as it isfound in the ‘General Provisions’ applicable to all obligations, not just contractual obliga-tions. Moreover, it explicitly governs all the stages in the life of the obligation, not just itsexecution. See above n 60. Judicial and doctrinal interpretations of the French Art 1134 (3)CC have likewise extended the ambit of that disposition to cover all stages of the life ofcontractual obligations. See, eg, Terré et al, above n 20, at §§ 177, 414.

63 Maleville, above n 40, at § 453, fn 23, citing Boyer, Gaudemet, Mazeaud andMazeaud, Flour and Aubert, Dupichot, Dereux, Chabas, Lopez Santa Maria and Salle de laMarnière.

64 H Beale, ‘Exclusion and Limitation Clauses in Business Contracts: Transparency’ inBurrows and Peel, above n 2, at 191, 200; Ghestin, above n 53, at § 407; Delebecque andMazeaud, above n 61, at 388–90.

65 Terré et al, above n 20, at § 297 (discussing the Cour de cassation jurisprudence of thesecond half of the nineteenth century, in which clauses providing for the payment of steepprofessional fees in contracts for professional services were almost systematically struckdown); Delebecque and Mazeaud, above n 61, at 363 (‘[the courts] having considered nullconventions of non-liability, precisely for moral reasons and by reason of the fundamentalvalue attached to the idea of fault, …’); Cass civ 2e 17 February 1955 D 1956.17 (‘are nullthe clauses excluding or limiting liability in delictual matters, given that articles 1382 and1383 of the Civil Code are of public order and their application hence not possibly paralyzedby a convention’).

66 Terré et al, above n 20, at § 456.




the first is designated as ‘subjective’,67 ‘real’68 or ‘explicative’69 interpreta-tion whereas the second is respectively referred to as ‘ideal’, ‘divinatory’ or‘creative’ interpretation, among other pairs of contrasting labels.70 Suchinsistence is consistent with the different locations in the code of the rulesthat govern these two phases respectively. The disciplining rules of articles1134 and 1135 CC, being found among the Dispositions générales ofChapter III, are kept separate from the consecrating rules pertaining to theclassic canons of interpretation contained in the De l’interprétation desconventions section. The strict delineation of the consecrating and disci-plining functions, just like the overall dominance of the latter, is thereforereflected in both phases of the interpretive stage.

Hence the view of contract law that emerges from the French materialson contractual interpretation is that of an overall disciplining frameworkwith a pocket of consecrating, which remains small and subject to thediscipline of the larger framework. The conception of contractual intentionas actual, empirical intention which grounds this pocket accounts for itsbeing both ultimately subjected to the discipline of the larger frameworkand nonetheless clearly delineanable. As an empirical matter, subjectivecontractual intention cannot but draw its normative significance from thesurrounding legal framework, which in turn sets the terms upon whichsuch significance is to be supplied. Whereas these terms are merelyevidentiary in the first, ‘deciphering’ phase of the interpretive process, theyalso are substantive in the second, ‘processing’ phase. But subjectiveintention as an empirical matter also exhibits sufficient conceptual tight-ness to sustain the integrity of the pocket as against the larger framework.In that sense, the view of contract law that emerges from the civil lawmaterials on contractual interpretation can be described as that of aprocess in which the disciplining function is far more prominent than theconsecrating one, yet the two functions remain neatly delineated from oneanother.

III. (ENGLISH) COMMON LAW MATERIALS ON CONTRACTUALINTERPRETATION

In contrast to the civilian view of contractual interpretation, the view ofcontract law that emerges from the common law materials is that of aprocess in which the consecrating and disciplining functions are not clearly

67 Maleville, above n 40, at § 452.68 Flour and Aubert, above n 27, at § 396.69 Terré et al, above n 20, at § 325.70 Pineau et al (above n 27, at 989) use the labels ‘explicit’ v ‘implicit’ interpretation,

which is perhaps less apt as the first phase includes an investigation of explicit and implicitparty intention.

94 Catherine Valcke



delineated, more specifically, in which the disciplining function tends to befolded into the consecrating function in an attempt to downplay theformer’s relative significance. I propose to show that this is the case byraising, and ultimately dismissing, some potential counter-examples, thatis, by identifying the rules and institutions of English law that most appearto embody a clear consecrating/disciplining distinction and/or a disciplin-ing dominance, and then demonstrating that they do so in appearance only.These rules and institutions are the jurisdictional division between law andequity as it pertains to contracts; the law/fact distinction; the doctrine ofimplied terms; the particular treatment of onerous contractual terms; andthe doctrine of frustration and certain elements of the law of remedies.

The jurisdictional division between law and equity—probably the mostfundamental organising feature of English private law—at first sight seemsas if it might supply a consecrating/disciplining delineation similar to thatwhich animates French contract law. Under that division, the courts of lawgenerally were to enforce contracts as expressed by the parties, whereas thecourts of equity were to set the contract aside where justice would demandit.71 If only from a jurisdictional perspective, therefore, the consecratingand disciplining functions seemingly were to be neatly compartmentalised.

The law/equity division never really tracked the consecrating/discipliningdistinction elaborated in Part I, however. For the contractual intention thatthe law courts have had to enforce is the parties’ objective intention72—theintention ‘which the party in question by his actions or words displays tothe other, not some hidden intention which he may have concealed in theinner reaches of his mind’.73 And as suggested, objective intention is arelatively loose concept, at least in comparison with subjective intention.At the outset, the very objective/subjective distinction at English law iselusive at best. Although objective and subjective elements are clearly bothat play in the English conception of contract (albeit with a clear predomi-nance of the former), their respective domains have yet to be reasonablyneatly delineated.74 Most significantly, however, the ‘objective’ intention of

71 See the chapter entitled ‘Of the Equitable Jurisdiction in Relieving Against Unreason-able Contracts or Agreements’ in Powell on Contracts (1790), cited in S Waddams, The Lawof Contracts, 4th edn (Toronto, Canada Law Book, 1999) 319.

72 See Blackburn J’s famous statement in Smith v Hughes (1871) LR 6 QB 597, 607 (‘If,whatever a man’s real intention may be, he so conducts himself that a reasonable man wouldbelieve that he was assenting to the terms proposed by the other party, and that other partyupon that belief enters into the contract with him, the man thus conducting himself would beequally bound as if he had intended to agree to the other party’s terms’). More recentrestatements include: Lord Reid in McCutcheon v David MacBrayne Ltd [1964] 1 WLR 125(HL) 127; Lord Diplock in Ashington Piggeries Ltd v Christopher Hill Ltd [1972] AC 441(HL) 502; Lord Steyn in Deutsche Genossenschaftsbank v Burnhope [1995] 1 WLR 1580(HL) 1587 [Burnhope].

73 Abella and Rothstein JJ in Double N Earthmovers, above n 2, at [65].74 For a suggestion that contract formation pertains to the subjective domain, whereas

contract interpretation relates to the objective domain, see: SA Smith, Contract Theory




English law itself appears to combine both kinds of elements. Englishcourts indeed commonly appeal to the shady notion of ‘the parties’reasonable intention’, without specifying whether by that they mean ‘whatthe parties reasonably ought to have intended’ or ‘what the partiesreasonably ought to be taken to have intended’. The difference betweenthese two meanings is critical, as the first betrays a clear discipliningstance, whereas the second can be seen as strictly consecrating. Yet Englishcourts have been reluctant to clarify which of these they mean, arguablybecause they in fact have been meaning both: the parties’ ‘reasonableintention’ stands for the intention which it is reasonable for them to haveprecisely because that is the intention which it is reasonable for each ofthem to attribute to the other. In other words, it is because a particularintention reasonably can (factually) be attributed to the parties that thecourt will endorse that intention as that which reasonably can (legally) beattributed to them. Hence does it make sense for an English court to affirmin one and the same breath that ‘a release cannot apply, or be intended toapply to circumstances of which a party had no knowledge at the time heexecuted it’.75 The same interplay of factual and normative—of consecrat-ing and disciplining—is reflected in such statements as: ‘[t]he purpose [ofinterpretation] is to enable the Court to attribute the appropriate objectivemeaning to the words used by the parties in the document’.76 And it isbecause of this interplay, I would argue, that contractual intention asobjective intention is inherently unstable, and hence conceptually looserthan subjective intention.

The conceptual looseness of objective intention arguably is part of theexplanation for the traditional prominence of literalism and the parol

(Oxford, Oxford University Press, 2004) 174; K Lewison, The Interpretation of Contracts(London, Sweet & Maxwell, 2004) 28–31. Recent case law on contractual interpretationhowever veers dangerously close to the subjective domain. See in particular Lord Hoffmann’srecent suggestion that it is only for reasons of ‘practical policy’ that the parties’ subjectiveknowledge is excluded from the matrix of facts which judges ought to consider wheninterpreting contracts (Investors Compensation Scheme Ltd v West Bromwich BuildingSociety et al [1998] 1 WLR 896 (HL) 913 [ICS]). See also Lord Nicholls, ‘My Kingdom for aHorse: The Meaning of Words’ (2005) 121 LQR 577, 586–9 (pleading for the admissibilityof previous negotiations and declarations of intentions for the purpose of interpretingcontracts); and Rix J (as he then was) in BHP Petroleum Ltd v British Steel plc ([1999] 2Lloyd’s Rep 583 (QBD), affirmed in part: [2000] 2 Lloyd’s Rep 277 (CA) (suggesting that theHadley v Baxendale test was ‘conceptually difficult’ in so far as it entails that where a lossnormally falling under the second branch of the test is in fact known to the defendant itbecomes knowable by a reasonable person in the position of the defendant, and thusrecoverable under the first branch of the test).

75 Pollock CB in Lyall v Edwards (1861) 158 ER 139 (QB) 143 (emphasis added).76 Hobhouse LJ in Credit Lyonnais Bank Nederland NV v Export Credit Guarantee

Department [1998] 1 Lloyd’s Rep 19 (CA) 38 (emphasis added). Another example is LordSteyn’s statement that ‘[t]he purpose of interpretation is to assign to the language of the textthe most appropriate meaning which the words can legitimately bear’. Equitable LifeAssurance Society v Hyman [2000] 1 AC 408 (HL) 458 [Hyman].

96 Catherine Valcke



evidence rule in English contract law—the other part pertaining to suchnotorious institutional factors as the endeavour to keep contractual inter-pretation matters away from the (often illiterate) juries.77 Although neitherliteralism nor the parol evidence rule is logically entailed by objectiveintention,78 their emergence is hardly surprising in a context whereunconstrained judges have historically been disapprovingly equated withlegislating judges and certainty in transactions is generally highly prized.As the notion of objective intention, in its looseness, would not provide therequisite certainty and constraining power, these would be suppliedthrough extraneous rules, for example, courts ought not look beyond ‘thefour corners of the document’ when interpreting contracts, and wordsought to be given their narrowest, most literal meaning.79 Just as the ruleof precedent arose as an effective means by which to curb the judges’legislative leanings and to restore certainty, so literalism and the parolevidence rule emerged as effective means tailored for contractual interpre-tation.

In turn, the need arose for a doctrine along the lines of rectification tocorrect the abuses generated through blind adherence to literalism and theparol evidence rule.80 Hence, one could exceptionally bypass the fourcorners of the document as well as the literal meaning of the words in caseswhere exclusive reliance on them would be unfair since neither reflects theparties’ actual intention. If in some cases hardship or injustice may beeffected by this rule of law, such hardship or injustice can generally beobviated by the power in equity to reform the contract, in proper cases andon proper evidence that there has been a real intention and a real mistakein expressing that intention; these matters may be established, as theygenerally are, by extrinsic evidence. The court will thus reform or rewritethe clauses in order to give effect to the real intention. However, that is notconstruction, but rectification.81

In the context of contractual interpretation centred upon objectiveintention, therefore, the purportedly respectively consecrating and disci-plining roles of law and equity are at times reversed: interpretation of thedocument at law plays the consecrating role where the document reflects

77 Pioneer Shipping Ltd v BTP Tioxide Ltd [1982] AC 724 (HL) 736 (Lord Diplock)[Pioneer Shipping].

78 In strict logic, it indeed is possible to say, as the French do, that literalism and the parolevidence rule pertain to the kind of evidence that is deemed useful for the purpose of gettingto contractual intention however substantively defined.

79 J Steyn, ‘Written Contracts: To What Extent May Evidence Control Language?’ (1988)41 Current Legal Problems 23; LH Hoffmann, ‘The Intolerable Wrestle with Words andMeanings’ (1997) 114 South African Law Journal 656; Lewison, above n 74, at 11.

80 See generally J Steyn, ‘The Intractable Problem of the Interpretation of Legal Texts’(2003) 25 Sydney Law Review 5, 8; Nicholas, above n 1, at 47–8; A Burrows, ‘Constructionand Rectification’ in Burrows and Peel, above n 2, at 77, 78.

81 Lord Wright in Inland Revenue Commissors v Raphael [1935] AC 96 (HL) 143.




the parties’ intention, whereas alteration of the document at equity playsthat role where the document is not reflective of party intention.82

Conversely, it is possible to describe both these operations as playing adisciplining role: where the court favours an objective interpretation of thecontract that demonstrably diverges from the parties’ subjective intention,it can be said that the court imposes its own interpretation upon theparties, and where the court instead looks beyond the document to reachto the parties’ subjective intention, it likewise can be said that the court isthen reading into the contract the words ‘which it considers ought to havebeen used’.83 If anything, therefore, the law/equity distinction cuts across,rather than tracks, the consecrating/disciplining distinction elaborated inPart I.

Law and equity have since formally merged, and the ripple effect is stillbeing felt in all fields of English law. That merger contributed little in theway of disentangling the consecrating and disciplining functions, however,arguably because contractual interpretation remains anchored in objectiveintention. Along with the gradual disappearance of juries in private lawmatters and the concurrent consolidation of the rules of law and equity,strict adherence to literalism and the parol evidence rule was eventuallyovercome.84 The parol evidence rule has come to be considered as morethan just a rule of evidence: ‘the so-called parol evidence rule … [is not] arule of evidence (though, like all legal rules, it has evidentiary conse-quences): it is a rule of substantive contract law, namely, that extrinsicstatements do not affect the parties’ obligations’.85 Whereas a similarfusion of the evidentiary and the substantive would, as discussed, be highlyunlikely in a legal system based on subjective intention, it is to be expectedin one that centres upon objective intention. For under the objectiveconception of intention, evidence of intention is, as between the contract-ing parties, in many respects as significant as intention itself: the onlyrelevant intention is that of which the other party has been given evidence.

82 ‘[B]oth techniques are concerned to ensure that the contract, as enforced, reflects theintentions of the parties. The construction of a contract achieves this by interpretation of thewords used. Rectification achieves this by alteration of the words used.’ Burrows, above n 80,at 77.

83 R Calnan, ‘Construction of Commercial Contracts: A Practitioner’s Perspective’ inBurrows and Peel, above n 2, at 17, 19 (criticising Lord Hoffmann in ICS, above n 74.)

84 ‘[I]n the case of commercial contracts, the restriction on the use of background hasbeen quietly dropped.’ Lord Hoffmann in Mannai Investment Co Ltd v Eagle Star LifeAssurance Co Ltd [1997] AC 749 (HL) 779. ‘Accordingly all the reasonably available andrelevant background information is now admissible for the purposes of construing private lawdocuments.’ G McMeel, ‘The Principles and Policies of Contractual Construction’ in Burrowsand Peel, above n 2, at 27, 45. For a recent Canadian relaxation of the parol evidence rule, seeGallen v Allstate Grain Co Ltd (1984) 9 DLR (4th) 496 (BCCA).

85 Waddams, above n 71, at 225–6. To the same effect: R Stevens, ‘Objectivity, Mistakeand the Parol Evidence Rule’ in Burrows and Peel, above n 2, at 101, 107: ‘[P]roperlyunderstood it is not, or at least is no longer, a rule of evidence but a rule of construction.’

98 Catherine Valcke



In that sense, the inherent conceptual looseness of objective intentioncarries over to the qualification of the parol evidence rule, as it does moregenerally to any line-drawing exercise as between evidentiary and substan-tive matters. Thus the parol evidence rule, just like objective intention, farfrom exemplifying our consecrating/disciplining distinction, does in factrun counter to it:

the ‘parol evidence’ rule is not, as it is sometimes portrayed, a rigid rule basedupon a now unfashionable love of certainty which will lead to the frustration ofthe parties’ actual intentions. Rather the effect of the rule is to enforce, notfrustrate, the intentions of the parties as objectively manifested.86

As for the movement away from literalism, it likewise appears to havecarried with it the conceptual looseness of objective intention. Having longstruggled from under the literalist legacy, English courts nowadays appearto be generally agreed that contractual words ought to be given their‘natural and ordinary meaning’.87 In contrast with the literal approach, thecurrent ‘natural and ordinary meaning’ approach—also known as ‘contex-tual’, ‘purposive’, ‘commonsensical’ or ‘commercial’88—considerably wid-ens the ‘matrix of facts’ which judges ought to consider when interpretingcontracts, to the point that ‘absolutely anything which would have affectedthe way in which the language of the document would have beenunderstood by a reasonable man’ ought now be included.89 Just like thenotion of objective intention, ‘natural’ meaning appears to combine avariety of judicial meanings90 among which are the linguistically naturalmeaning (the meaning of dictionaries),91 and the legally natural meaning

86 Stevens, ibid, at 108.87 ICS, above n 74; Bank of Credit and Commerce International v Ali [2002] 1 AC 251

(HL) [BCCI] (whether an employee releasing ‘all claims’ against an employer in anemployment termination agreement covers claims not known to exist in law at the time ofsigning the agreement.)

88 McMeel, above n 84, at 40.89 ICS, above n 74 at 913. The relation between the consolidation of law and equity and

the movement towards greater interpretive flexibility is vividly illustrated by that case. LordHoffmann’s extension of the matrix of facts to be considered in interpreting contracts went sofar as to blur the line between general principles of contractual interpretation and theequitable claim of rectification. The blurring is particularly striking in the following passage,highly reminiscent of rectification yet meant as a comment on contractual interpretationgenerally: ‘[If it seems] from the background that something must have gone wrong with thelanguage, the law does not require judges to attribute to the parties an intention which theyplainly could not have had.’ Lord Hoffmann in ICS. For more detailed discussions, seeCalnan, above n 83, at 19; Burrows, above n 80, at 77.

90 J Carter and E Peden, ‘The “Natural Meaning” of Contracts’ (2005) 21 Journal ofContract Law 277. Carter and Peden list three different meanings, not just two, although it isnot entirely clear how the third, the ‘application to the contract’ meaning, differs from thefirst two: McMeel, above n 84, at 31.

91 For example, Lord Mustill in Charter Reinsurance Co Ltd v Fagan [1997] AC 313(HL) 384 (‘ [the meaning of the words] in the sense of their primary meaning in ordinaryspeech’). The House of Lords there unanimously rejected the argument that the words




(the meaning of precedents).92 Also, as with the notion of objectiveintention, judges seem content to remain elusive as to which of the twomeanings of ‘natural’ they are appealing,93 presumably for the same reasonthey do not view these meanings as clearly distinct. The general movementtowards greater interpretive flexibility therefore did not contribute todisentangling the consecrating and disciplining functions any more thandid the drift of the parol evidence rule towards substantive law. Insummary, so long as contractual interpretation remains centred upon theinherently unstable notion of objective intention, the consolidation of lawand equity and the attendant liberalisation of contractual interpretation isunlikely to help clarify the line between the consecrating and discipliningfunctions of English contract law.

Nor is the fact/law distinction of English law likely to be more helpful inthis regard. Unlike at civil law, where the line between the consecrating anddisciplining functions is reinforced through the different labels (subjective/ideal, explicative/creative and so on) and designations (issues of fact/law)assigned to the two phases of the interpretive stage, the task of determiningthe normative content of a given contract at English law is indifferentlylabeled ‘interpretation’ or ‘construction’,94 and treated as a question of lawregardless of its being primarily driven by consecrating or disciplining

‘actually paid’ in a reinsurance treaty could be taken to entail that the (then insolvent)reinsured had to demonstrably discharge his liability to the underlying insured beforeclaiming from the reinsurer.

92 ‘[T]he legal effect … of the linguistic meaning which results from the application of therules and principles of contract law to the linguistic meaning …’ Carter and Peden, above n90, at 279–80. See, eg, Lord Wilberforce in Prenn v Simmonds [1971] 1 WLR 1381, 1385(‘The words used may, and often do, represent a formula which means different things to eachside, yet may be accepted because that is the only way to get ‘agreement’ and in the hope thatdisputes will not arise. … [In such a case, one must] try to ascertain the ‘natural’ meaning.’).

93 Carter and Peden, above n 90, at 279–80. See, eg, Pagnan SpA v Tradax OceanTransportation SA ([1987] 1 All ER 81 (QBD), where Steyn J found that the ‘naturalmeaning’ of the words used by the parties in a contract for the sale of tapioca pellets forimportation into the then EEC was ‘more consistent’ with a strict obligation on the part of thesellers to provide the requisite export certificates than with merely an obligation of reasonablediligence on their part. The Court of Appeal ostensibly agreed ([1987] 3 All ER 565 (CA)),but then proceeded to apply a default rule to determine the legal (as opposed to linguistic)effect of the words. That rule impliedly applied because it derived from the applicable law andhad not been explicitly waived by the parties. ‘Ultimately, the conclusion was one of lawbased on precedent rather than ‘natural’ meaning.’ Carter and Peden, above n 90, at 281.

94 ‘The expression “construction”, as applied to a document, at all events as used byEnglish lawyers, includes two things: first the meaning of the words; and secondly their legaleffect, or the effect which is to be given to them.’ Lindley LJ in Chatenay v BrazilianSubmarine Telegraph Co Ltd [1891] 1 QB 79 (CA) 85. See also McMeel, above n 84, at 32;Carter and Peden, above n 90, at 279. Sometimes ‘interpretation’ is used as a subset of‘construction’ to designate interpretation of the express terms of the contract, in opposition tothe ‘implication’ of terms into an agreement: Lord Steyn in Hyman above n 76, at 408, 458–9(CA and HL).

100 Catherine Valcke



motivations.95 At the same time, it is widely agreed that ‘[t]he meaning ofan ordinary word of the English language is not a question of law’.96

Inevitably, therefore, the question of law that is the interpretation ofcontracts nonetheless is one that cannot but be interspersed with questionsof fact.97 Doubts have even been raised as to whether the designation ofcontractual interpretation as a question of law is applicable beyond therealm of written contracts.98 Here again, the ambivalence arguably reflectsthe interplay of factual and normative inherent in contractual intention asobjective intention. In light of the normative dimension of objectiveintention, the interpretive process could hardly be designated as merely‘explicative’ and a question of fact. At the same time, the designation as‘normative’ and a question of law is somewhat misleading given theincontrovertible factual rooting of that intention, as already explained.

The discussion of Part III thus far has aimed to expose the difficulty ofdisentangling the consecrating and disciplining functions at English law.But the entanglement of these functions is not symmetrical. It is not thecase, in other words, that these functions are interwoven into one anotherso as to cause the resulting fabric to be ‘neutral,’ namely, neither domi-nantly consecrating nor dominantly disciplining. Rather, as we suggested atthe outset, the disciplining function tends to be folded into the consecratingfunction in an apparent attempt to downplay the former’s relative signifi-cance. That is to say, the consecrating function is put forward as dominant,yet various strands of disciplining at times emerge from behind that front.The contrast with French law thus lies, not merely in the difficulty indisentangling the two functions, but also in the overall consecrating ratherthan disciplining dominance, which likewise can be traced to the ground-ing notion of objective intention.

As explained, objective intention combines the respectively factual andnormative dimensions of the two functions. As a particular conception ofcontractual intention, however, it itself is first and foremost a consecratingnotion. However much disciplining may be at play within the notion ofobjective intention, therefore, it remains disciplining that is nestled withina consecrating shell. Thus, whereas at French law contractual interpreta-tion unfolds as a disciplining drama with a cameo appearance by a

95 As per Lord Diplock in Pioneer Shipping, above n 77, and in Bahamas InternationalTrust Co Ltd v Threadgold [1974] 1 WLR 1514 (HL).

96 Lord Reid in Cozens v Brutus [1973] AC 854 (HL) 855 (there commenting on an issueof statutory interpretation).

97 ‘Although the ascertainment of the meaning of a written contract is a question of law,many steps in the process of ascertaining that meaning are classified as questions of fact.’Lewison, above n 74, at 96.

98 See Romer LJ’s suggestion that the interpretation of an oral contract is ‘entirely aquestion of fact’, Torbett v Faulkner [1952] 2 TLR 659 (CA) 661.




consecrating character, at English law it showcases a consecrating charac-ter struggling to suppress its disciplining streak.

Take the issue-of-law designation just discussed. The fact that contrac-tual interpretation is so designated despite objective intention also contain-ing a factual dimension could be taken as a statement that contract lawought to be viewed as primarily disciplining in purpose. As this designationrecently was emphatically reiterated outside the context of the jurysystem,99 it indeed could reasonably be viewed as a somewhat deliberateaffirmation of the courts’ self-perception as a disciplining body withrespect to matters of contractual interpretation. As one English judgedeclared, in an unmistakably disciplining tone, ‘[when interpreting acontract] the question to be answered always is, ‘What is the meaning ofwhat the parties have said?’ not ‘What did the parties mean to say?’ … itbeing a presumption juris et de jure … that the parties intended to say thatwhich they have said.’100

A closer look at the ‘natural meaning’ approach which judges adopt forthe purpose of answering that question however, reveals that the nodtowards the disciplining function in fact is only half hearted. LordHoffmann himself views the ‘natural meaning’ as resulting from a mergerof the legal into the linguistic, that is, from a process whereby the legal ismade to conform to the linguistic, not the other way around. In his dissentin Bank of Credit and Commerce International v Ali, he indeed exclaimed:

[i]f interpretation is the quest to discover what a reasonable man would haveunderstood specific parties to have meant by the use of specific language in aspecific situation at a specific time and place, how can that be affected byauthority?101

Further in the same case:

[T]he general trend in matters of construction … has been to try to assimilatejudicial techniques of construction to those which would be used by a reasonablespeaker of the language in the interpretation of any serious utterance in ordinarylife … ‘Almost all the old intellectual baggage of ‘legal’ interpretation has beendiscarded.102

99 Pioneer Shipping, above n 77.100 Lord Simon in Schuler (L) AG v Wickman Machine Tool Sales Ltd [1974] AC 235

(HL) [Schuler], reading from Norton on Deeds (1906), 43.101 BCCI above n 87, at [51]. ‘How can the question of what a reasonable man in 1990

would have thought BCCI and Mr Naeem meant by using the language of an ACAS form beanswered by examining what Lord Keeper Henley said in 1758 (Salkeld v Vernon (1758) 1Eden 64, 28 ER 608)?’ Ibid. Lord Bingham likewise said (at [9]) of the ‘natural meaning’ thatit ‘seems to me to be both good law and good sense: it is no part of the court’s function tofrustrate the intentions of contracting parties once those have been objectively ascertained’(emphasis added).

102 Ibid, at [62], reiterating his comments in ICS, above n 74, at 912.




But the court’s ruling in that case is even more interesting than itsstatements. The Court of Appeal had adopted a narrow reading of thecontract and ruled against the employee on the question of construction,but had then ruled in his favour on the second question, whether the Bankon equitable grounds should be entitled to rely on the contract.103 TheHouse of Lords in contrast adopted a broader, ‘natural meaning’ readingof the contract, which led it to dismiss the Bank’s appeal without having toconsider the equity question. The ‘natural meaning’ approach thus pro-vided sufficient flexibility for the House of Lords to do, under the headingof legal construction, what the Court of Appeal had considered it couldonly do under the heading of equity. The ‘natural meaning’ approach thusallowed the House of Lords to do all the disciplining it felt it had to do yetpackage its decision as a primarily consecrating operation.

To be sure, the ‘natural meaning’ of a contract is to be determined partlyin light of the ‘nature and object of the contract’,104 which in turn inprinciple is for the court largely to determine objectively, as a matter oflaw.105 Just as the classification of the contract in accordance with its objetis considered an issue for objective determination at French law, therefore,its ‘characterisation’ at English law is determined, if not exclusivelyobjectively, at least ‘not [as a] pure question of construction’.106 Thus is the‘natural’ meaning of a commercial contract the ‘commercially appropriate’meaning,107 to the point where ‘if detailed semantic and syntacticalanalysis of words in a commercial contract is going to lead to a conclusionthat flouts business commonsense, it must be made to yield to businesscommonsense’.108

Closer inspection of the reasons offered by English courts in such caseshowever reveals that these courts, more than their French counterparts, arereluctant to see themselves as speaking ‘for the law’ as opposed to ‘for theparties’. Unlike at French law, the very ‘object’ of the contract at Englishlaw often is presented in terms suggestive of what the parties considered

103 [2000] 3 All ER 51, (HL) 69.104 Darlington Futures Ltd v Delco Australia Pty Ltd (1986) 161 CLR 500, (HCA) 510

(‘[from] the context in which the clause appears including the nature and object of thecontract’).

105 ‘The commercial or business object of a provision, objectively ascertained, may behighly relevant: … But the court must not try to divine the purpose of the contract byspeculating about the real intention of the parties.’ Lord Steyn in Burnhope, above n 72, at1587 (emphasis in original).

106 McMeel, above n 84, at 36. For a clear articulation of the difference betweencontractual ‘construction’ and ‘characterisation’ at English law, see: Street v Mountford[1985] AC 809 (HL).

107 Prenn v Simmonds, above n 92 (‘profit’ given the most commercially sensiblemeaning).

108 Antaios Compania Naviera SA v Salen Rederierna AB (The Antaios) [1985] AC 191(HL) 201 (time charterparty; shipowners allowed to withdraw vessel ‘on any breach of thischarterparty’; interpreted as ‘on any repudiatory breach’).




this object to be. In a recent case concerning an insurance policy containinga clause excluding liability for negligence, for example, Lord Hoffmannstated, in support of the clause’s enforceability, that

[n]egligence is a risk which the parties could reasonably have been expected toallocate to one party or the other, so as best to achieve the commercial objectivesof the contract.109

The Supreme Court of Canada likewise declared, in a recent case involvinga municipal contract issued by tender, that

[t]he rationale for the tendering process, as can be seen from [the tender]documents, is to replace negotiation with competition. … But M.J.B. Enterprisesmakes clear that the tender documents control the contractual obligations of theparties to a tender, and Iacobucci J’s observations were based on the particulardocuments in that case.110

The issue of the proper determination of the object of the contractmoreover has arisen in other, equally significant contexts, for example incases of mistake of identity. In the famous case of Ingram v Little, alljudges agreed that the issue of whether the buyer’s identity was an essentialelement of a sale agreement ought to be determined by reference to theobject of the agreement, but disagreed as to how one determines thatobject.111 The majority took the view that the buyer’s identity was essentialto the agreement merely because the seller had at the outset made that clearto the buyer.112 Lord Devlin dissented, stating that the object of theagreement was to be derived from its nature as impartially determined bythe court, not from what the parties in fact intended it to be.113 On thatbasis, Lord Devlin would have concluded that the buyer’s identity, whilepresumably possibly essential to a loan agreement for example, could notbe essential to the kind of agreement there at issue, namely, a saleagreement. Lord Devlin’s view arguably114 was recently confirmed by theHouse of Lords in Shogun Finance Ltd v Hudson, where Lord Hobhousedetermined that, the agreement at bar being a consumer credit agreement,the identity of the rogue there had to be considered an essential elementregardless of the parties’ own views on the matter.115 Interestingly, this

109 Lord Hoffmann in HIH Casualty & General Insurance Ltd v Chase Manhattan Bank[2003] 2 Lloyd’s Rep 61 (HL) [67] [HIH Casuality].

110 Abella and Rothstein JJ in Double N Earthmovers, above n 2, at [57], quoting fromMJB Enterprises Ltd v Defence Construction (1951) Ltd [1999] 1 SCR 619, 170 DLR (4th)577 at [41] (emphasis added; citations omitted).

111 [1961] 1 QB 31.112 Sellers LJ, ibid, at 51.113 Ibid, at 68.114 Stevens, above n 85, at 112.115 [2004] 1 AC 919 (HL). See in particular Lord Walker at [185].




determination caused quite a stir in the English legal community,116 whichis itself revealing, I would suggest, as to the hold that the consecratingfunction retains over the English legal psyche.117 In sum, the consecratingdominance of the ‘natural meaning’ approach is palpable even in its appealto the object of the contract.

But what of contract terms that have traditionally eluded the applicationof the general principles of interpretation just described? English courtshave at times acknowledged that clauses considered particularly onerouscall for greater disciplining. Hence the tighter interpretive frameworks putin place to govern, in particular, penalties, forfeitures, clauses in restraintof trade, exclusionary clauses and releases. Some such clauses—forfeitures,118 penalties,119 clauses in restraint of trade120—have beenconsidered so onerous as to warrant strict ‘rules of law’ against theirenforcement, rules whereby such clauses would be unenforceable regard-less of the parties’ implicit or explicit intention. Other clauses—exclusionary clauses121 and releases122—have, not without somehesitation,123 been determined as less onerous yet sufficiently exacting towarrant the establishment of ‘rules of construction’, whereby enforcementis disallowed unless otherwise explicitly provided by the parties.

However, whereas the reasons offered in support of the establishment ofrules of law admittedly are often framed in clear disciplining terms,124

116 See, eg, A Phang, P-W Lee and P Koh, ‘Mistaken Identity in the House of Lords’[2004] CLJ 24; C MacMillan, ‘Mistake as to Identity Clarified?’ (2004) 120 LQR 369; CHare, ‘Identity Mistakes: A Missed Opportunity?’ (2004) 67 MLR 993; K Scott, ‘MistakenIdentity, Contract Formation and Cutting the Gordian Knot’ [2004] Lloyd’s Maritime andCommercial Law Quarterly 292; G McMeel, ‘Interpretation and Mistake in Contract Law:The Fox Knows Many Things …’ [2006] Lloyd’s Maritime and Commercial Law Quarterly49.

117 It admittedly is difficult to tell whether the outcry was primarily directed at LordHobhouse’s suggestion that the contractual object is to be determined objectively or at thefact that he used the parol evidence rule to make that determination, as this distinctionapparently eludes many commentators. Stevens, above n 85, at 107–10.

118 See, eg, Vernon v Bethell (1762) 2 Eden 110, 28 ER 838 (forfeiture clause in mortgagecontract).

119 See, eg, Protector Endowment Loan and Annuity Co v Grice (1880) 5 QBD 592 (CA).[Grice].

120 See, eg, A Schroeder Music Publishing Co Ltd v Macaulay [1974] 1 WLR 1308 (HL)[Schroeder v Macaulay].

121 Photo Production Ltd v Securicor Transport Ltd [1980] AC 827 [Photo Production].122 BCCI, above n 87. In BCCI, the court explicitly applied the law of exclusionary

clauses.123 George Mitchell Ltd v Finney Lock Seeds Ltd [1983] QB 284 (CA) [George Mitchell],

affirmed [1983] 2 AC 803 (HL); BCCI, above n 87.124 ‘The law itself will control that express agreement of the party; and by the same reason

equity will let a man loose from his agreement, and will against his agreement admit him toredeem a mortgage.’ Howard v Harris (1683) 1 Vern 190, 23 ER 406, 407; ‘[N]ecessitousmen are not, truly speaking, free men, but, to answer a present exigency, will submit to anyterms that the crafty may impose upon them.’ Vernon v Bethell, above n 118, at 839.Similarly with respect to penalty clauses: ‘[E]quity in truth refused to allow to be enforced




those offered to support rules of construction are more ambivalent. Thevery fact that the courts ultimately settled for the rule-of-constructiondesignation with respect to exclusionary clauses arguably confirms theconsecrating preference of English courts, given that ample fairness (disci-plining) grounds were available to justify the alternative, rule-of-lawdesignation.125 One possible (disciplining) reason for the courts nonethe-less settling upon the rule-of-construction designation admittedly is theirbeing satisfied that, absent any formal finding of unconscionability, theunfairness of the situation is either neutralised or trumped where theparties have expressly contracted around it. It is arguably more likely,however, that the courts favoured that designation because rules ofconstruction involve only minimal judicial intervention with the contract-ing process in so far as these rules only require of the parties that they beexplicit as to their intention. This is borne out by such ubiquitous judiciallanguage as ‘[t]he Court has no right, in my opinion, to make a differentcontract for the parties’.126

That the courts’ preference for the rule of law designation was motivatedmore by a concern to minimise judicial intervention than by any real sensethat such designation better meets the demands of fairness is also borne outby recent declarations to the effect that the degree of judicial interventionentailed by rules of construction, however minimal, may still be consideredexcessive. Endorsing Lord Denning’s comments in George Mitchell Ltd vFinney Lock Seeds Ltd,127 Lord Hoffmann in BCCI indeed advocated thatsuch rules be altogether abandoned, and the ‘general principles of interpre-tation’ extended even to contract terms as onerous as releases, on theground that courts use rules of construction as cloaks behind which to hidewhen effectively rewriting contracts which they find to be unfair.128

Whether the switch to the ‘natural meaning’ approach in all cases will in

what was considered to be an unconscientious bargain.’ Grice, above n 119, at 596. And withrespect to clauses in restraint of trade: ‘if one looks at what [judges] said in the light of whatthey did, one finds that they struck down a bargain if they thought it was unconscionable asbetween the parties to it, and upheld it if they thought it was not. So I would hold that thequestion to be answered … is: ‘Was the bargain fair?’ Lord Diplock in Schroeder v Macaulay,above n 120, at 1315. Admittedly, institutional reasons also are at play here: Waddams, aboven 71, at [449].

125 ‘[E]xemption clauses … often amounted to taking with one hand what had been givenwith the other … [a] contracting party undertook various obligations and then provided thathe was not to be liable if he failed to perform them.’ Lord Hoffmann in BCCI, above n 87, at[66].

126 ‘allocating the risk of misfeasance or non-feasance … in a manner different from theallocation of that risk that the clause . . . envisages.’ Lord Scott, dissenting, in HIH Casualty,above n 109, at 126.

127 George Mitchell above n 123, at 296–7.128 ‘When judges say that “in the absence of clear words” they would be unwilling to

construe a document to mean something, they generally mean (as they did in the case ofexemption clauses) that the effect of the document is unfair.’ Lord Hoffmann in BCCI, aboven 87, at [61].




fact result in greater transparency is far from clear in light of thecontortions which the courts have, as suggested, long inflicted upon thatexpression. More probably, courts will continue to cling to reasonscouched in consecrating language,129 despite the fact that these all are caseswhere ‘[n]o amount of scrutiny of the words will provide an answer thatcan be said to be based solely on the expressed intention of the parties’.130

Nonetheless, the move from rule of construction to the general ‘naturalmeaning’ approach, just like that from rule of law to rule of construction,itself is significant as it confirms that the courts tend to feel morecomfortable speaking purportedly for the parties than openly for the law.

Such packaging of disciplining as consecrating moreover appears toextend to onerous clauses beyond just exclusionary clauses and releases.Where French judges, as indicated, regularly strike all sorts of contractterms on no other ground than that these terms qualify as ‘abusive’ at law,English judges typically prefer to interpret onerous terms away, to find thatthese terms are not really a part of the parties’ agreement, as ‘[t]he moreunreasonable the result the more unlikely it is that the parties can haveintended it’.131 These are just particular instances of a general pattern that,in the opinion of several observers, sees unconscionability (that is, disci-plining) concerns underlying many more contract decisions than thedeciders care to acknowledge.132 This pattern perhaps is nowhere moreobvious than in Lord Denning’s notorious, repeated and failed attempts toreverse it.133

129 See, eg, the suggestion that the Canada Steamship rules of interpretation of clausesexcluding or limiting liability for negligence (cited in E Peel, ‘Wither Contra Proferentem?’ inBurrows and Peel, above n 2, at 53, 60) are ‘merely guidelines to assist the court inascertaining the true intentions of the parties.’ Peel, ibid, at 61, citing Lamport and Holt Linesv Coubro and Scrutton (M & I) Ltd (The Raphael) [1982] 2 Lloyd’s Rep 42 (CA). Elsewherethese rules have similarly been described as resting on the premise that it is ‘inherentlyimprobable that one party … should intend to absolve the other party from the consequencesof the latter’s own negligence.’ Gillespie Bros & Co Ltd v Roy Bowles Transport Ltd [1973]QB 400 (CA) at 419 (Buckley LJ). See also: Lord Wilberforce in Ailsa Craig Fishing Co Ltd vMalvern Fishing Co Ltd [1983] 1 WLR 964 (HL) 966 (‘The relevant words must be given, ifpossible, their natural, plain meaning.’)

130 Carter and Peden, above n 90, at 282–3 (emphasis in original).131 Lord Reid in Schuler, above n 100, at 251 (emphasis added), cited in McMeel, above n

84, at 43. Strikingly, McMeel describes this statement as ‘a candidate for the most importantprinciple of construction.’ Ibid.

132 ‘Several generations of common lawyers have been educated in the belief that thecommon law of contracts admits no relief from contractual obligations on grounds ofunfairness, or inequality of exchange. … It will be suggested, however, that the law ofcontracts, when examined for what the judges do, as well as for what they say, shows thatrelief from contractual obligations is in fact widely and frequently given on the grounds ofunfairness, and that general recognition of this ground of relief is an essential step in thedevelopment of the law.’ Waddams, above n 71, at 319. See more generally, Smith’s discussionof the realist interpretation of consideration: Smith, above n 74, at 228ff.

133 Solle v Butcher [1950] 1 KB 671 (CA) (with respect to mistake); Lloyd’s Bank vBundy [1975] 1 QB 326 (CA) (with respect to unconscionability generally); George Mitchell,




The merger of the disciplining and consecrating functions and domi-nance of the latter is also reflected in the judicial treatment of implicitcontract terms. The implicit content of the contract at English law dividesinto terms implied in fact and terms implied by law,134 the latter apparentlyrepresenting ‘the minimum decencies … which a court will insist upon asessential to an enforceable bargain of a given type’.135 Here again itaccordingly may look as if English law mimics French law in offering aclear consecrating/disciplining distinction, one in which the discipliningfunction is unswerving and fully acknowledged. But that once again ismore apparent than real. To begin with, the place of implied terms in theoverall law of contract is not the same at English law and at French law:‘[t]he French starting-point is that the incidents of a contract are fixed bylaw, subject to the parties’ power to vary them … [whereas] the startingpoint for a Common law court is the implied term’.136 The very fact thatEnglish lawyers do through implied terms (the law of the parties) whatFrench jurists do through rules (the law of the State) is revealing as to theselegal actors’ respective conceptions of contract law as primarily consecrat-ing and disciplining.137 In addition, the distinction between terms impliedin fact and terms implied by law is bound to be murkier at English law, inso far as the English notion of a term implied by law, just like that ofobjective intention, embodies within it a combination of fact and law. Thefollowing passage from Lord Hoffmann’s extra-judicial writings on impliedterms indeed is highly reminiscent of our remarks concerning objectiveintention:

In fact, of course, the implication of a term into a contract is an exercise ininterpretation like any other. It may seem odd to speak of interpretation when,by definition, the term has not been expressed in words, but the only difference

above n 123 (with respect to exclusionary clauses); Stockloser v Johnson [1954] 1 QB 476(CA) (with respect to forfeiture clauses); British Movietonews Ltd v London & DistrictCinemas Ltd [1951] 1 KB 190 (CA) 201 (with respect to frustration-like events).

134 ‘General default rules’ and ‘ad hoc gap fillers’ according to Lord Steyn in Hyman,above n 76, at 458–9. Examples of term implied by law would be the implied terms of theimplied duty of trust and confidence in the employment relationship in Mahmud v Bank ofCredit and Commerce International SA [1998] AC 20 (HL) and Johnson v Unisys Ltd [2003]1 AC 518 (HL) [35] (Lord Hoffmann). On implied terms generally, see: G Treitel, The Law ofContract, 11th edn (London, Sweet & Maxwell, 2003) 201–13.

135 K Llewellyn, Book Review of O Prausnitz, The Standardization of CommercialContracts in English and Continental Law (1938–39) 52 Harvard Law Review 700, 703(emphasis added).

136 Nicholas, above n 1, at 49. Nicholas then pursues (at 49–50): ‘Where French lawbegan with the Roman system of typical contracts and superimposed on it the unitaryconsensual principle that any agreement is a contract, English law reversed the process. In sofar as it thinks in terms of typical contracts, it has derived them from the general principle ofcontract through the device of the implied term.’

137 ‘[English legal actors] cling to the legal fiction that the courts do not make contracts forthe parties. All must be traced back to the implicit will of the parties’ David and Pugsley,above n 1, at 264.




is that when we imply a term, we are engaged in interpreting the meaning of thecontract as a whole. For this purpose, we apply the ordinary rule of contractualinterpretation by which the parties are depersonalised and assumed to bereasonable.138

When describing the two kinds of implied terms at English law, itaccordingly is more appropriate to refer to a gradual ‘process of thehardening of fact into law’139 or ‘shades on a continuous spectrum’,140

than to the kind of strict qualitative distinction at play at French law. Forthe same reason, moreover, the realm of terms implied in fact is necessarilycomparatively larger at English law: it includes all terms qualifying asproxies for the parties’ reasonable intention, unlike at French law, where itincludes only the terms that qualify as proxies for the parties’ actualintention. English courts indeed have correspondingly tended to interpret‘terms implied at law’ narrowly, as ‘terms implied at statutory law’, unliketheir French counterparts, for whom ‘law’ there means the law of generalapplication (the Code) as well as the law of exception (statutes).141 ButEnglish courts have constricted the realm of terms implied by law evenfurther by confining their application to cases of strict ‘necessity’, whereasterms implied are in fact instead subjected to the looser test of ‘businessefficacy’.142 Finally, the justifications produced in support of terms impliedby law often are embedded in strongly consecrational language. LordDiplock himself once said of these terms that they flow from a ‘presump-tion … that the parties by entering into the contract intended to accept theimplied obligations’.143 Just like terms implied in fact, therefore, termsimplied by law apparently ultimately draw their legitimacy from havingbeen (implicitly) intended by the parties. Despite its apparent similaritywith its French counterpart, therefore, the English law distinction betweenterms implied by law and terms implied in fact turns out to be much lessclear, and its disciplining dimension much more equivocal.

Other instances of disciplining packaged as consecrating can be foundbeyond the realm of ‘interpretation’ proper, finally. The law of frustration

138 Hoffman, above n 79, at 662.139 Nicholas, above n 1, at 50. Nicholas gives the examples of implied terms relating to the

sale of goods and to partnership, which began as reasonable implication, later becamepresumptions juris de jure, and were eventually codified.

140 Lord Wilberforce in Liverpool City Council v Irwin [1977] AC 239 (HL) 254.141 See, eg, the Supreme Court of Canada’s classification of the implied terms deemed

included in tendering contracts as terms implied in fact (flowing from ‘the presumedintentions of the parties’) rather than as terms implied by law (‘the legal incidents of aparticular class or kind of contract’), despite the existence of case law establishing that suchterms were to be implied in tendering contracts. Double N Earthmovers, above n 2, at [30].

142 E Peden, ‘Policy Concerns Behind Implication of Terms in Law’ (2001) 111 LQR 459.The stricter test of ‘necessity’ admittedly recently has been considerably relaxed: McMeel,above n 84, at 32.

143 Photo Production, above n 121, at 850. See also Lord Diplock referring to LordWilberforce at 851.




provides one example; the law of remedies another. Neither of these legaldomains is alluded to in the present survey of French materials for thesimple reason that it would never come to the mind of a French jurist toqualify either of these as bearing in any way upon contractual interpreta-tion. The doctrine of imprévision—the French counterpart of the doctrineof frustration— is generally viewed as an issue that, just like that ofcontract formation, pertains to the objective delimitation of the contractualrealm and accordingly cannot possibly be determined by the parties’ ownviews on the matter.144 Likewise with the law of remedies, it would neveroccur to a French jurist to think of the sanctions attached to the violationof contractual obligations as issues of interpretation: when such obliga-tions are violated, it is for the law to intervene and apply whateversanction is deemed objectively warranted.145 By the Cartesian logic so dearto the French, the parties cannot possibly be both the sanctioned and thesanctioners. The justification most commonly offered in support of the rulelimiting damages recoverable for breach to those flowing ‘immediately anddirectly’ from the breach,146 indeed, appeals to objective notions ofcausation, not subjective notions of party intention.147 The rules sanction-ing violations of contractual obligations, like the doctrine of imprévision,therefore, at French law are viewed as clearly falling on the discipliningside of the consecrating/disciplining divide.

Not so at English law. To be sure, frustration and the law of remediesare, just like their French counterparts, typically classified under rubricsother than contractual interpretation. Moreover, one does find judicial148

and academic149 accounts of these as disciplining matters, which accountsvery much resemble the French descriptions of imprévision and thesanctions des obligations. Here end the similarities, however. With respectto frustration, English courts have long deployed creative interpretation

144 See generally Terré et al, above n 20, at §§ 445ff.145 As two prominent French jurists remarked with respect to exclusionary clauses: ‘the

contractual freedom to determine the consequences of a failure to live up to one’s contractualobligations ends where the essence of the contract begins.’ Delebecque and Mazeaud, above n61, at 378–9 (emphasis in original).

146 Art 1151 CC; Art 1613 CCQ.147 See, eg, Terré et al, above n 20, at §§ 567ff; Pineau et al, above n 27, at § 464.148 See, eg, Cotton LJ in Hydraulic Co Ltd v McHaffie (1878) 4 QBD 670 (CA) 677: ‘It

cannot be said that damages are granted because it is part of the contract that they shall bepaid: it is the law which imposes or implies the term that upon breach of a contract damagesmust be paid.’

149 See, eg, Stevens, above n 85, at 105–107: ‘This process does not depend upon thecourt’s ability to imply positively as a matter of fact a condition precedent (‘if the music halldoes not exist there is no contract’) or a condition subsequent (‘if the music hall burns downthe agreement comes to an end’) into the agreement. In many cases such implication will bewholly artificial as it cannot be determined what the parties would have agreed if they hadthought about the matter in advance. Rather the court’s task is the negative one of construingthe limits of the actual bargain the parties have entered into ….’




(consecrating) arguments in order to shirk the issue altogether.150 And incases where the doctrine has seemed inescapable, they have openly resistedapplying it, while at times proving overly generous in their interpretationof force majeure clauses.151 It has been suggested that the reason for suchdisparate treatment is that a finding of frustration causes ‘the contract [tobe] immediately and automatically brought to an end, irrespective of thewishes of the parties’,152 whereas enforcing force majeure clauses incontrast involves ‘no danger of becoming involved in making a newcontract for the parties or imposing an outcome irrespective of theirwishes’.153 Likewise with remedies, the rationale for the remoteness rule ofdamages, whether in its first inception in Hadley v Baxendale154 or in itsmultiple subsequent iterations, is very much consecrational in tone, at leastwhen compared to the causation rationale favoured by the French. Thistone is particularly strong in the so-called ‘second contract theory’ cases, ofwhich Horne v The Midland Railway Co is a prime example.155 BlackburnJ there famously affirmed that ‘in order that the notice [of the plaintiff’s

150 See, eg, the House of Lords’ judgment in British Movietonews Ltd v London andDistrict Cinemas Ltd [1952] AC 166 (HL). The fact that the implied contract theory offrustration was altogether abandoned in the second half of the twentieth century confirmsthat such a consecrating twist on an unambiguously disciplining issue was too artificial to besustainable. I am grateful to Stephen Waddams for this point.

151 See, eg, J Lauritzen AS v Wijsmuller BV (The Super Servant Two) [1990] 1 Lloyd’s Rep1 (CA). (carriage of goods—contract provides that carrier will use one of two barges—forcemajeure clause allowing for cancellation in the event of ‘perils or danger and accidents of thesea’—carrier performance rendered impossible as one barge sank and other was allocated toperformance of another contract—carrier’s frustration argument denied as their inability toperform primarily due to decision to allocate second barge to other contract—carrier’s forcemajeure argument allowed (despite same reasoning applicable: impossibility to perform dueto decision concerning second barge rather than ‘perils of the sea.’))

152 E McKendrick, ‘Force Majeure Clauses: The Gap between Doctrine and Practice’ inBurrows and Peel, above n 2, at 233, 239, paraphrasing the Privy Council in Hirji Mulji vCheong Yue Steamship Co Ltd [1926] AC 497 (PC) 505, 509 (emphasis added).

153 Ibid.154 ‘Where two parties have made a contract which one of them has broken, the damages

which the other party ought to receive … should be such as … may reasonably be supposed tohave been in the contemplation of both parties, at the time they made the contract. Now, ifthe special circumstances under which the contract was actually made were communicated bythe plaintiffs to the defendants, and thus known to both parties, the damages resulting fromthe breach of such a contract … would be the amount of injury which would ordinarilyfollow from a breach of contract under these special circumstances so known and communi-cated…. For, had the special circumstances been known, the parties might have speciallyprovided for the breach of contract by special terms as the damages in that case; and of thisadvantage it would be very unjust to deprive them.’ Hadley v Baxendale (1854), 9 Exch 341,156 ER 145, 151 (Alderson B) (emphasis added). Ironically, the court there heavily reliedupon the writings of the famous French jurist Pothier! This is not the only time that anEnglish court clumsily undertook to graft a French doctrine upon the English law. For asimilar initiative with respect to contractual mistake, see: R David, ‘La doctrine de l’erreurdans Pothier et son interprétation par la common law d’Angleterre’ in P Matter (ed), Étudesde droit civil à la mémoire de Henri Capitant (Paris, Dalloz, 1939) 145.

155 (1873) LR 8 CP 131 (Exch Ch) 135 (emphasis added).




particular circumstances] may have any effect, it must be given under suchcircumstances, as that an actual contract arises on the part of the defendantto bear the exceptional loss’.156

It is worth emphasising that the difference between French and Englishlaw here again is rhetorical, not functional. Functionally, the two systemsdiffer little, in so far as they both allow for the possibility of contractingaround the rules: force majeure and liquidated damages clauses indeed arepermitted and commonly used in both systems.157 With respect to bothsystems, therefore, it can plausibly be asserted that, in the absence of suchclauses, the parties ought to be taken to have implicitly endorsed the rulesof objective law—the doctrines of frustration/imprévision, the rules ofremoteness/dommages direct et immédiat. From a rhetorical rather than afunctional perspective, however, what is interesting is that the ubiquity offorce majeure and liquidated damages clauses in practice has much affectedthe perception of the corresponding rules of objective law at English law,but not at French law. English lawyers have, in light of this practice, tendedto see the rules of objective law as mere ‘default rules’, which can bewaived at will,158 and which accordingly bear at least some connection tointerpretation issues. French jurists in contrast have quarantined suchdepartures to the realm of (consecrating) ‘exceptions’ so as to prevent themfrom contaminating the (disciplining) ‘principles’ of objective law. TheEnglish tendency to blur the disciplining/consecrating line and recast thelatter as the former, and the reverse French tendency to emphasise thedisciplining/consecrating line and acknowledge the former’s dominance,are thus palpable even beyond the realm of contractual ‘interpretation’proper.

The view of contract law that emerges from the English materials oncontractual interpretation hence is that of a process in which the disciplin-ing function tends to be folded into the consecrating function in anapparent attempt to downplay its relative significance. The difficulty indisentangling the two functions originates from the fact that objectiveintention—the cornerstone of the English contract law—itself combinesconsecrating and disciplining elements, and the resulting conceptual loose-ness carries over to the various legal instruments pertaining directly (and

156 Ibid, at 141. See also Transfield Shipping Inc v Mercator Shipping Inc [2008] UKHL48, discussed in A Robertson, ‘The Basis of the Remoteness Rule in Contract’ (2008) 28 LegalStudies 172.

157 With respect to liquidated damages, see Arts 1152–3 CC and Art 1622 CCQ. Withrespect to force majeure clauses, see: Mazeaud et al, above n 27, at § 581; Art 1470 CCQ.

158 This may explain why the ‘default rule’ explanation of objective law propounded bylaw and economics scholars has been better received in common law than in civil lawjurisdictions. The easier reception of law and economics scholarship in common lawjurisdictions arguably is another element that confirms the dominance of the consecratingfunction in the common law psyche, as that scholarship generally aims to justify legal rules byappeal to individual preferences.




less directly) to contractual interpretation. The historical division betweenlaw and equity, the subsequent convergence of law and equity andattendant liberalisation in the interpretation of written contracts, thelaw/fact distinction, the doctrine of implied terms, the particular treatmentof onerous terms, and even such institutions as the doctrine of frustrationand certain elements of the law of remedies indeed show legal actorsmerging the two functions. This merger is far from symmetrical, however,in that English judges clearly prefer to see themselves as consecrating(speaking for the parties) rather than disciplining (speaking for the law)agents. The consecrating dominance of the grounding notion of objectiveintention carries over to the judicial treatment of contractual object andonerous clauses, as well as to the doctrine of frustration and the remote-ness rule of damages. In light of the difficulty in disentangling the twofunctions, the pocket metaphor used for the purpose of describing therelation of subordination of the consecrating to the disciplining at Frenchlaw hardly is apposite for describing the reverse relation at English law.The better metaphor would be ‘a front of consecrating dominance fromwhich strands of disciplining emerge from time to time’.

IV. CONCLUSION

As an exercise in comparative legal rhetoric, our analysis has focused uponthe justifications, that is, the reasons offered in support of the particularchoices of legal instruments deployed for the purpose of interpretingcontracts at French and English law. For those reasons embody the corevalues that animate each system, in particular, their respective conceptionsof contract law. The conception of contract law that emerges from theFrench materials on contractual interpretation is, largely owing to theconceptual tightness of the grounding notion of subjective intention, thatof a process in which the disciplining function is by far the more prominentyet the two functions remain neatly delineated from one another. Incontrast, the conception of contract law that emerges from the Englishmaterials is, owing to the conceptual looseness of objective intention, thatof a process wherein the two functions are interwoven and the significanceof the disciplining function is comparatively downplayed. Although thesedifferent conceptions may well be more or less ‘right’ or ‘good’ or‘desirable’ from some extra-legal perspective, be it economic, philosophicalor otherwise, all that matters for now is that they be internally accurate:that they correctly reflect the distinctive self-understandings that animateFrench and English contract law respectively.

As the reconstruction of these different conceptions simultaneouslyserved to interconnect the materials on contractual interpretation in eachsystem and to contrast those of the other system, I hope that this article




contributed to a deeper understanding of the rules and institutions at issue,that is, an understanding of these rules and institutions as they relate toeach other and to those in the other system. This is what is meant above bya ‘meaningful comparative understanding of law’.




5

Consideration and the Morality ofPromising

ANDREW S GOLD*

I. INTRODUCTION

WHEN IS IT problematic for contract law to diverge frompromissory morality? Seana Shiffrin has made a significantcontribution to contract theory debates by focusing attention on

this question. In a recent article, she supports an intermediate, ‘accommo-dationist’ approach to the relationship between law and morality, anapproach which suggests that the law should be responsive to moralconcerns without actively enforcing interpersonal morality as such.1

According to Shiffrin, moral agents should not be placed in a positionwhere they have to accept a conflict between the justifications for legaldoctrine and their moral beliefs. Applying this framework, however,Shiffrin concludes that US contract law and promissory morality improp-erly diverge, and she offers the consideration doctrine as a primaryexample of this divergence. She critiques contract law because promisorshave a moral duty to keep their word whether or not there is considera-tion, yet contract law only regards promises as enforceable when consid-eration is provided or the promisee has reasonably relied on a promise tohis or her detriment.2

* I would like to thank Curtis Bridgeman, Richard Bronaugh, Michael Pratt, SongRichardson and Stephen Siegel, as well as participants in a faculty workshop held atDePaul University, for their helpful comments. Any errors are my own.

1 See S Shiffrin, ‘The Divergence of Contract and Promise’ (2007) 120 Harvard LawReview 708.

2 Ibid, at 709–10.



This article adopts Shiffrin’s general accommodationist approach, butchallenges her application of that approach to the consideration doctrine.3

Accommodationists are concerned with the virtuous moral agent’s abilityto accept the justifications for legal doctrine, given that agent’s interest inacting morally. The key legal structure in the contract setting is the privateright of action: contract law functions through private rights of actionwhich depend on a promisee’s decision to bring suit. This legal structureimplicates the morality of the promisee.

Notably, an individual promisee may not be morally justified in coercingthe promisor in cases of breach (whether through legal or non-legalmeans). The virtuous moral agent’s understanding of the private right ofaction is accordingly significant under an accommodationist approach.Moral agents—from the perspective of a promisee—can be expected to seetheir legal power to initiate litigation as a means to remedy violations oftheir moral rights, and they may understand the law’s justification in theseterms. The convergence or divergence of contract law and morality shouldtherefore be assessed with respect to the morality of a promisee’s initiationof a private right of action against a promisor.

Given this backdrop, an accommodationist approach requires a determi-nation of when it is morally acceptable for a promisee to seek coerciveremedies if a promisor breaks his or her word. This inquiry calls intoquestion Shiffrin’s conclusions about the consideration doctrine. In somecases, such as promises for bargained-for consideration, it may be morallyacceptable for the promisee to coerce the promisor based on a failure toperform.4 In other cases, morality will conflict with such coercion. Consid-eration doctrine is plausibly understood as a legal response to moralconcerns about coercive responses to a breach—by refusing to enforcegratuitous promises, courts can avoid enabling promisees to indirectlycoerce promisors when morality would conflict with a promisee’s efforts tocoerce. As a result of differences in the morality of coercive remedies fordistinct types of promises, the consideration doctrine may then represent aconvergence with promissory morality, and not a troubling divergence.This article suggests that, under reasonable assumptions about promissorymorality, the consideration doctrine is actually consistent with accommo-dationist principles.

3 As noted, this article focuses on the consideration doctrine. Charles Fried has recentlyraised doubts about other doctrinal areas in which Shiffrin suggests a divergence exists. See CFried, ‘The Convergence of Contract and Promise’ (2007) 120 Harvard Law Review Forum1.

4 For example, the presence of consideration may justify a claim of ownership ofcontractual performance by the promisee. See generally A Gold, ‘A Property Theory ofContract’ (2009) 103 Northwestern University Law Review 1 (suggesting a normativeconnection between the presence of consideration and a promisee’s right to insist oncontractual performance).

116 Andrew S Gold



II. THE ACCOMODATIONIST APPROACH TO CONTRACT ANDMORALITY

In her recent article, Shiffrin contends that contract law diverges from themorality of promising, and that this divergence is a cause for concern.Before reviewing specific areas of divergence, it is worth considering whysuch a divergence is relevant. Shiffrin calls for a reassessment of howprivate law and interpersonal morality are, and should be, related. In herview, a divergence places an unacceptable burden on moral agents whoseek both to meet the law’s obligations and also to comply with interper-sonal morality.

Shiffrin suggests that there are two primary strands of thought oncontracts and promises. One approach, which she calls the ‘reflective’approach, ‘take[s] interpersonal morality as a template for legal rules’.5

The justifications for a reflective approach vary. Supporters often acceptthis perspective based on the perceived nature of law, or for reasonsgrounded in their political philosophy. In either case, this view seeks tomove legal content in the direction of morality. An alternative view, the‘separatist’ approach, suggests that law and morality should be viewed asindependent spheres. Some separatists have ‘normative, political reasons’for their opposition to legally forcing people to comply with rules ofinterpersonal morality.6 Others simply believe that law, or contract law inparticular, has distinct goals from interpersonal morality—for example, thepurpose of contract law may be the establishment of an efficient system ofexchange.7

As Shiffrin indicates, both accounts have advantages to them. However,she concludes that these two approaches do not exhaust the availablepossibilities. Her insight is to develop another option for assessing therelationship between morality and law, one which borrows insights fromreflective and separatist accounts. This approach takes an intermediate,‘accommodationist’ position on the proper relationship between moralityand law. Shiffrin suggests that ‘even if enforcing interpersonal morality isnot the proper direct aim of law, the requirements of interpersonalmorality may appropriately influence legal content and legal justificationsto make adequate room for the development and expression of moralagency’.8

Part of the reason for Shiffrin’s accommodationist position is herrecognition that moral agents believe that ‘there are moral duties to obey

5 Shiffrin, above n 1, at 713.6 Ibid.7 Ibid.8 Ibid, at 715.

Consideration and the Morality of Promising 117



the law’.9 This raises a divergence concern because the law may end upregulating activity which is also subject to distinct moral norms. In light ofthis possibility, Shiffrin argues that ‘legal rules should be sensitive to thedemands placed on moral agents so that law-abiding moral agents do not,as a regular matter, face substantial burdens on the development andexpression of moral agency’.10 Contracts, moreover, are a good example ofhow moral agents may simultaneously participate in legal and moralspheres. Contracts clearly impose significant legal duties, and Shiffrinconcludes that promises are embedded in contracts, such that contractsalso implicate the moral duties intrinsic to promising.11

Based on these premises, Shiffrin derives the following principle:

[W]hen a legal practice is pervasive and involves simultaneous participation in amoral relationship or practice, the content and normative justification for thelegal practice must be acceptable to a reasonable moral agent with a coherent,stable, and unified personality.12

In light of this principle, Shiffrin also proposes three criteria against whichlegal rules should be measured. First, she suggests that legal rules ‘shouldnot, as a general matter, be inconsistent with leading a life of at leastminimal moral virtue’.13 Contract law is not typically thought to violatethis first criterion. The second criterion is that ‘the law and its rationaleshould be transparent and accessible to the moral agent’.14 This limitationis more subtle in its operation, and if adopted, its significance for contractlaw is potentially large. Following this criterion, the justifications profferedfor the law’s substance should, if accepted, be compatible with theindividual’s ‘developing and maintaining moral virtue’. Moral agents mustbe able to both understand and accept the law’s justification as their own.I will refer to this as the ‘transparency’ criterion.15 Lastly, Shiffrin contendsthat ‘the culture and practices facilitated by law should be compatible witha culture that supports morally virtuous character’.16 This cultural crite-rion does not require the law to enforce interpersonal morality as such. Infact, legal interference in certain moral areas may be counterproductive.Even so, some legal doctrines, such as doctrines designed to facilitateefficient breaches of contract, might tend to undermine morally virtuous

9 Ibid.10 Ibid.11 Ibid, at 721.12 Ibid, at 717.13 Ibid, at 718.14 Ibid.15 I follow Michael Pratt’s lead in using this terminology. Pratt refers to this as the

‘transparency constraint.’ See M Pratt, ‘Contract: Not Promise’ (2008) 35 Florida StateUniversity Law Review 801, 805.

16 Shiffrin, above n 1, at 719.

118 Andrew S Gold



behaviour.17 To the extent a moral culture is harmed by legal doctrine, thisthird, cultural criterion comes into play.

The key development in Shiffrin’s argument is her suggestion that oneneed not endorse a reflective approach to legal doctrine in order to betroubled by a divergence between the law and moral principles. One can bedeeply sceptical about a legal system that mandates moral behaviour assuch, yet still remain concerned with the compatibility between the normsexpressed in legal doctrine and the norms commonly accepted by moralagents. Accommodationism allows legal institutions to address this com-patibility problem without forcing individuals to comply with any particu-lar comprehensive conception of moral virtue.

I do not intend to challenge Shiffrin’s accommodationist perspective inthese pages. For the purposes of this article, it will be assumed that herframework is appropriate. Nor will I question her view that promises areembedded in contracts and form their basis.18 Instead, this article willassess which aspects of promissory morality should be relevant to anaccommodationist approach. This analysis, in turn, will cast doubt onShiffrin’s claim that the consideration doctrine diverges from promissorymorality.

III. THE PURPORTED DIVERGENCE BETWEEN PROMISSORYMORALITY AND THE CONSIDERATION DOCTRINE

There are several aspects of contract law that Shiffrin offers as examples ofa divergence between contract law and morality. These include the consid-eration doctrine, the use of expectation damages rather than specificperformance, mitigation requirements, and the absence of punitive dam-ages.19 This article leaves most of these potential divergences to one side.Instead, it questions the purported divergence between promissory moral-ity and the consideration doctrine. In the process, this article also suggeststhat the relationship between promissory morality and contractual enforce-ment is more complex than Shiffrin’s argument implies.

As a starting point, it will be helpful to explain the basis of Shiffrin’scritique of contract law and the consideration doctrine. She suggests thatcontract law diverges from the morality of promises because, despite theassociation of contract obligations with morally binding promises, the

17 For Shiffrin’s argument to this effect, see ibid, at 740–43.18 Cf Pratt, above n 15 (critiquing Shiffrin’s comparison of contracts and promises).19 Some of these claims are controversial. For further analysis of these other claims, see

Fried, above n 3.




resulting legal obligations do not correspond to the moral obligations ofpromisors.20 She then uses the consideration doctrine as an example ofsuch a divergence:

For instance, the moral rules of promise typically require that one keep aunilateral promise, even if nothing is received in exchange. By contrast, contractlaw only regards as enforceable promises that are exchanged for something or onwhich the promisee has reasonably relied to her detriment.21

Shiffrin’s analysis thus focuses on the moral requirement that one keep apromise—that is, it focuses on the moral duty of the promisor to thepromisee. From that perspective, the divergence appears to be straightfor-ward, for the promisor has a moral duty to perform regardless of thepresence or lack of consideration extending from the promisee. Legaldoctrine, in contrast, will often preclude a successful suit if the promisewas not made for consideration.22 From Shiffrin’s perspective, promissorymorality and contract law differ as to ‘whether unilateral promises bind’.23

In addition, even if the law does not on its face require citizens to acceptimmoral premises as justification for legal doctrine, a divergence betweencontract law and morality could negatively affect the development andmaintenance of a moral character. The existing law might encourageindividuals ‘to associate the conditions of binding agreements with quidpro quo exchange’.24 Under an accommodationist approach, the existenceof legal doctrine that is either premised on ideas which conflict withcommonly held views of interpersonal morality, or that is corrosive ofmoral character, poses a challenge. Shiffrin argues that the considerationdoctrine raises these concerns, in light of how moral agents understandcontracts.

IV. THE PROBLEM OF PROMISSORY RIGHTS AND REMEDIES

As the above discussion reveals, the content of contract law can be framedto suggest a divergence between law and morality. But it should be notedthat Shiffrin’s claim is not that we should derive contract law directly from

20 Shiffrin, above n 1, at 709.21 Ibid, at 709–10.22 Shiffrin also recognises the possibility that distinctively legal grounds may support a

divergence between contract and promissory morality. One potential source of distinctivelylegal grounds, which Shiffrin rejects as a sufficient explanation for the consideration doctrine,would involve the impact of legal enforceability on gift promises. As she notes, there is anargument that gifts do not serve their purpose if they are subject to mandatory enforcement.Ibid, at 736–7. For further analysis of this question, see M Eisenberg, ‘The World of Contractand the World of Gift’ (1997) 85 California Law Review 821, 846–52.

23 Shiffrin, above n 1, at 728.24 Ibid, at 742.

120 Andrew S Gold



the morality of promises—she does not endorse a ‘reflective’ approach.Instead, her thesis builds on ‘the law’s self-description’ that contracts ‘[rest]upon promises per se’.25 It calls for a reassessment of contract law basedon the extent to which contract law and promissory morality diverge,given the public understanding that contracts are promissory in nature.

Contract theories which purport to explain and justify contract law interms of promissory morality have been critiqued for failing to distinguishbetween juridical obligations and merely ethical obligations. For example,consider Charles Fried’s theory of contracts. Fried looks to promise-basedduties and the value of trust as a basis for contract obligations.26 This is areflective approach to the law. However, a reflective approach along theselines also has broad implications. Peter Benson has argued that, in order toexplain contract law: ‘Fried must hold that a breach of promise, as anabuse of trust, necessarily infringes a right in the promisee that can becoercively enforced.’27 Fried’s theory raises the question whether brokenpromises necessarily justify coercion.

Accommodationism does not call for contract law to reflect promissorymorality in all respects—the role of law under this view is not to enforcepromissory morality per se. In addition, accommodationism does not purportto explain how contract law results from that morality. In fact, Shiffrinexpressly avoids defining the purposes of contract law.28 Shiffrin’s project isconcerned with the proper relationship between contract law and promissorymorality. More specifically, she is concerned with the burdens that a virtuousmoral agent will face if the justifications for legal doctrine are inconsistentwith moral principles. The foundation of this concern is an apparent differ-ence between when promises bind morally and when contracts bind legally, inlight of contract law’s self-description in promissory terms.

Divergence as applied by Shiffrin has a specific meaning. Shiffrin locatesa divergence based on the degree to which the content and structure ofcontract law ‘parallel’ interpersonal morality. As she suggests:

Contract law would run parallel to morality if contract law rendered the sameassessments of permissibility and impermissibility as the moral perspective, exceptthat it would replace moral permissibility with legal permissibility and it would useits distinctive tools and techniques to express and reflect those judgments.29

25 Ibid, at 722.26 See C Fried, Contract as Promise: A Theory of Contractual Obligation 17 (Cambridge,

Harvard, 1981): ‘The moralist of duty thus posits a general obligation to keep promises, ofwhich the obligation of contract will be only a special case—that special case in which certainpromises have attained legal as well as moral force. But since a contract is first of all apromise, the contract must be kept because a promise must be kept.’

27 See P Benson, ‘The Idea of a Public Basis of Justification for Contract’ (1995) 33Osgoode Hall Law Journal 273, 292.

28 Shiffrin, above n 1, at 721.29 Ibid, at 722.




Thus, the absence of a legal duty where a moral duty is present triggers herview that a divergence exists.

The weakness in this theory is in the notion that the structure of contractlaw should ‘parallel’ promissory morality. Analogical reasoning is generallysubject to the argument that the similarities supporting an analogy may notbe the right similarities. Other similarities could always be chosen.30

Shiffrin’s argument falls prey to a comparable difficulty, depending uponwhich parallels are selected. Parallels between contract law and moralitycan be found in many ways. Thus, a necessary starting point when findingparallels between contract law and promissory morality is to determinewhich parallels are the ones which should matter.

Shiffrin’s understanding of when contract law parallels the world ofpromises selects legal and moral permissibility as decisive factors. Yet noguiding principle tells us which features of promissory morality should bethe ones which contract law must parallel.31 If selecting a different parallelbetween contract law and promissory morality would lead to conflictingresults, then the use of parallels will offer limited assistance in analysingcontract law. For example, what if the question is, does contract lawparallel promissory morality with respect to secondary rights (that is,remedial rights)? A search for parallels here suggests a different outcomefrom Shiffrin’s analysis. In her article, Shiffrin notes that legal remedies andmoral remedies may diverge.32 She appears to view a promisee’s moralremedies as a part of promissory morality, and this article will follow thesame approach.33 However, once the promisee’s secondary rights are takeninto account, it is far from clear that the consideration doctrine improperlydiverges from promissory morality.

Although theories of promising vary considerably, it is commonlyunderstood that they involve both rights and duties respecting perform-ance. These primary rights and duties are uncontroversial. The secondaryrights which flow from non-performance are much more contentious. A

30 On the difficulty in determining which similarities should matter in analogical reason-ing, see S Brewer, ‘Exemplary Reasoning: Semantics, Pragmatics, and the Rational Force ofLegal Argument by Analogy’ (1996) 109 Harvard Law Review 923, 932: noting that‘everything is similar to everything else in an infinite number of ways, and everything is alsodissimilar to everything else in an infinite number of ways’.

31 Although Shiffrin’s argument regarding consideration emphasises one parallel in par-ticular, however, Shiffrin does note that various factors would be in parallel if contract lawand morality were to ‘run parallel’. See Shiffrin, above n 1, at 722 n 24 (suggesting that‘[l]egal impermissibility would substitute for moral impermissibility, legal requirement formoral requirement, and so on’).

32 Ibid, at 724.33 This view of remedies is implicit in Shiffrin’s understanding, as she addresses judicial

orders of specific performance and monetary damages in terms of a divergence betweenpromise and contract. Ibid, at 722–4. If one considers the remedial question distinct frompromissory morality, the arguments which follow would remain valid as an accommodation-ist analysis of moral remedies.

122 Andrew S Gold



recent article by Margaret Gilbert on promissory obligations provides agood illustration of how these moral secondary rights are often subject tolimits.34 She argues that theories of promising should take into account theview that the promisee possesses a moral right to what was promised.35

Gilbert does not find, however, that promissory morality permits apromisee to physically force promisors to keep their word. As we will see,this remedial stopping point is significant. Standard intuitions about themorality of coercion can change with context, such as when bargained-forconsideration is present. If everyday, gratuitous promises may not becoercively enforced consistent with morality, and if promises withbargained-for consideration may be coercively enforced consistent withmorality, then the consideration doctrine offers a parallel between contractlaw and promissory morality.

A. The Moral Right to Performance of a Promise and its Significance forMoral Remedies

The discussion which follows will use a particular reading of promises toillustrate how a promisee’s moral rights can be significant to the accommo-dationist approach. Gilbert’s analysis suggests that the promissory duty toperform need not correspond to enforceable rights held by the promisee.She contends that a promisee has a moral right to performance, yet shequestions whether a promisee’s morally legitimate responses to breach ofan everyday promise include coercive force beyond a rebuke. That said,Gilbert’s focus in her discussion of moral rights is on ‘everyday’ promises.If we turn our attention to other types of promises, coercive remediesbecome more palatable. Certain promises, such as promises in exchangefor consideration, may justify stronger coercive remedies. This part of thearticle will thus suggest that morally acceptable remedies for promissorybreach may fall short of coercion, and in addition, that the acceptability ofcoercive remedies can vary from one type of promise to another. Acomparison of contract law and morality based on the bindingness ofcontractual promises therefore leaves out a potentially importantcomponent—an assessment of the morality of coercive remedies.

The idea of a ‘moral right’ provides a foundation for the discussion tofollow, and it plays a prominent role in Gilbert’s analysis of promising. Aswill become apparent, a ‘right’ in the contractual setting can have severalmeanings, covering both moral and legal concepts. The relevant concept inGilbert’s analysis is a moral concept, but in order to understand the idea of

34 See M Gilbert, ‘Scanlon on Promissory Obligation: The Problem of Promisees’ Rights’(2004) Journal of Philosophy 83.

35 Ibid, at 86.




moral rights, it is helpful to begin with legal understandings. Legaldiscourse regarding rights covers a variety of related ideas, ranging fromclaim rights to privileges, powers and immunities, and each are sometimesdenoted by the single word ‘right’. For clarity, it is helpful to use WesleyHohfeld’s terminology, which differentiates among these legal concepts.Under this rubric, a ‘claim right’ in the legal setting refers to those rightswhich correspond to legal duties.36 If party A has a legal claim right to aparticular action by B, then B necessarily has a correlative legal duty toperform this action. This idea of rights in terms of correlative duties is thecrucial one for present purposes.

As Herbert Hart has indicated, the correlation between claim rights andduties is more than a feature of law; it has a counterpart in the realm ofmorality.37 Drawing on Hart’s work, Gilbert recognises the existence ofnon-legal analogues to legal claim rights.38 These ‘moral rights’ involveclaims which one party has against another as a moral matter. Correspond-ing to the right holder’s claim, the other party in this moral relationshiphas a responsibility to provide what is due. If the individual with the dutyfails to provide what is due, this constitutes a wrong to the right holder.Promising is an example of a linguistic act which, if successful, ordinarilycreates such moral rights. The promisee’s moral rights are ‘directed’—theycorrespond to another party’s duties.39 Right holders possess their rightagainst others; the corresponding duty is directed toward the right holder.

Given these premises, there still remains an important point which needsclarification. The presence of a moral right indicates that the possessor ofthe right should be able to do something about the rights violation. At themost basic level, a right holder should be able to demand conductconsistent with the right. Assuming that promisees have moral rights, onemust determine the related question of what, if anything, the promisee canlegitimately do to enforce those rights. A common intuition about moralrights holds that where they exist, it is morally legitimate for the rightholder to seek to remedy a violation of their requirements.

For Hart, a moral right is linked to the ability to use force. As hesuggests:

The most important common characteristic of this group of moral concepts isthat there is no incongruity, but a special congruity in the use of force or the

36 This distinction is most often associated with the work of Wesley Hohfeld. ForHohfeld’s analysis of legal rights, see W Hohfeld, ‘Some Fundamental Legal Conceptions asApplied in Judicial Reasoning’ (1913) 23 Yale Law Journal 16. Gilbert likewise draws onHohfeld’s ideas in this area. See Gilbert, above n 34, at 87.

37 HLA Hart, ‘Are There Any Natural Rights?’ Philosophical Review, LXIV, 2 (April1955): 175–91.

38 See Gilbert, above n 34, at 87 (citing Hart, ibid).39 Ibid.

124 Andrew S Gold



threat of force to secure that what is just or fair or someone’s right to have doneshall in fact be done; for it is in just these circumstances that coercion of anotherhuman being is legitimate.40

Under Hart’s view, when one has a moral right to something, it islegitimate to limit the freedom of another. In some cases, the right holdermay even pursue a coercive remedy.41 Although Hart notes the possibilityof legitimate coercion, however, Hart’s above description of a moral rightdoes not distinguish between coercion in the form of self-help, andcoercion by means of legal mechanisms. The idea of a moral right ispotentially consistent with either form of coercion.42

Importantly, Gilbert suggests that ‘[c]ontemplating the exercise of“force” in connection with everyday promises and agreements may seem togo too far’.43 Although she believes promises create moral rights, herdescription of these moral rights raises doubts that they are congruent withthe use or threat of force beyond a rebuke.44 Everyday promises asdescribed by Gilbert apparently do not provide the promisee with afull-fledged authority to make others comply by physical force. Thepromisee has standing to demand performance, or even to rebuke someonewho fails in their duties, but more forceful responses are questionable.

Notice, however, that a moral right that supports a rebuke yet falls shortof permitting stronger coercive measures is substantially different from amoral right that supports the promisee’s use of force to make anotherperform their promise. Under commonly held intuitions about promising,if someone breaks an ordinary, gratuitous promise, the promisee canpublicly demand that he or she perform, and if the promisor fails to do so,

40 Hart, above n 37, at 178.41 Richard Bronaugh has suggested to me that this remedial aspect should be viewed in

terms of a promisee’s possession of a ‘power’. In the legal setting, a good example of aHohfeldian power would be the promisee’s ability to initiate a lawsuit against the promisor. Inthe non-legal setting, it is an interesting question whether a promisee’s response to a breachshould be referred to as the exercise of a moral power—say, as to make another liable tomoral criticism or release from moral responsibility. For a valuable contribution in this area,see J Raz, ‘Voluntary Obligations and Normative Powers’ in Supplementary vol XLVI,Proceedings of the Aristotelian Society 79, 92–6 (1972) (concluding that the concept of anormative power can be applied outside the law, and developing a theory as to when thisoccurs).

42 As will be developed in Part IV.B, both forms of coercion are significant whencomparing contract enforcement to promissory morality.

43 Gilbert, above n 34, at 89.44 It should be noted that Gilbert considers a rebuke to fall under the general category of

using ‘force’. See ibid, at 89 (‘It is reasonable … to include under the rubric of force the kindsof thing noted at the outset of this essay: such things as informal rebukes and demands.’). I donot share this view. However, whether or not a rebuke should count as ‘force’, there remainsan important distinction between a rebuke and the use of physical force, which Gilbert’sdiscussion highlights. Gilbert’s discussion indicates that one may believe that promises createa moral right to performance while doubting that it is moral to remedy a breach of aneveryday promise with physical force.




the promisee can rebuke the promisor for dishonesty. It is not usuallythought that the promisee may, consistent with morality, personally forcethe promisor to perform.

This question of whether forcible coercion is morally acceptable is alsosignificant to an accommodationist assessment of contract law. To theextent that contract law is justified in terms of promissory morality, therelevant legal doctrine may be grounded on the moral propriety of coerciveremedies as a response to breach, rather than the morality of keeping one’sword. In addition, the presence of consideration may be important indetermining whether coercive remedies are morally acceptable. Benson’sanalysis of contract law helpfully suggests this possibility.45 As Bensonindicates, legal doctrines like the consideration doctrine ‘suppose a distinc-tion between promises that create correlative rights and duties which arecoercible, and promises that may only give rise to an ethical duty offidelity’.46 If this distinction exists as a matter of promissory morality, thencontract law may parallel what promissory morality requires.

My own writing on contract theory offers support for a moral—and notsolely legal—distinction between promises based on bargained-for consid-eration. I have argued that a contractual promisee can acquire a propri-etary interest in a promisor’s future actions by meeting the terms of aconditional promise.47 This view of contracts builds on a Lockean concep-tion of property acquisition. Under this rubric, the doctrine of considera-tion is a basic feature of an acquisition process—when consideration hasbeen provided, a promisee will understand performance as belonging tohim or her. A promisee is able to acquire an enforceable right tocontractual performance because the act of providing consideration isanalogous to other settings in which an individual’s labour justifiesenforceable rights of ownership. Providing consideration according to apromisor’s terms can thus be a means of obtaining moral rights inperformance that are consistent with coercion in cases of breach.

Other theories of promising (and rights acquisition) may also supportthis conclusion. It is unnecessary for the reader to adopt a particular theoryof contracts, or even of promissory morality, to recognise the abovedistinctions regarding the implications of moral rights. Whatever thesource of our moral intuitions about coercive remedies, some promises arereasonably understood to support coercive remedies for breach, and otherpromises raise serious doubts in this regard.

Scepticism about exercising force—beyond a demand or rebuke—torespond to an ordinary promissory breach is widely shared. This suggests a

45 See Benson, above n 27. See also P Benson, ‘The Unity of Contract Law’ in P Benson(ed) The Theory of Contract Law (Cambridge, Cambridge University Press, 2001).

46 Benson, above n 27, at 293.47 Gold, above n 4.

126 Andrew S Gold



range of potentially legitimate responses to a breach of promise. A moralright to performance of a promise could mean the possessor of the righthas standing to demand the other party meet the obligation, and to rebukea failure to perform, or it could mean the possessor of the right hasstanding to demand performance and also, if performance is not forthcom-ing, to physically coerce the other party to meet the obligation. These aredistinct understandings of moral rights, and the remedies which thosemoral rights imply.

As shorthand, I will refer to the first example—even if it includesstanding to rebuke—as a ‘weak’ moral right, and the second example as a‘strong’ moral right.48 A weak moral right provides its owner with thestanding to demand what is owed, and even to rebuke someone who doesnot act consistently with their corresponding moral duty. A strong moralright provides its owner with the standing to force compliance with thecorresponding moral duty, or to otherwise remedy a failure to perform.Thus, a strong moral right could justify coercion by legal mechanisms, or itcould justify coercion by the right holder directly. Weak moral rights donot correspond to remedies that require physical force, while strong moralrights do so.49

As will be developed below, this distinction matters when assessing Shif-frin’s critique of contract law. If the morality of ordinary promises—promiseswhich lack consideration—does not include a strong moral right of the sortthat would support physical coercion as a remedy, then contract law does notdiverge from the morality of promising as far as Shiffrin suggests. A judicialunwillingness to legally enforce promises that lack consideration wouldparallel the moral impermissibility of a promisee using coercion to forceperformance in the absence of contract law. Furthermore, for contractualpromises with bargained-for consideration, the legal power to seek coerciveremedies in court may parallel the moral permissibility of using self-help toenforce such promises in the absence of contract law. The selection of parallelsbetween contract law and promissory morality thus has consequences. Theremaining question is whether the parallel Shiffrin emphasises—a parallelbased on the bindingness of promises—is nevertheless the correct parallel toemphasise given the likely perceptions of virtuous moral agents. The next partof this article addresses that question.

48 The range of possibilities is actually more elaborate than the two categories describedabove. A primary moral right could be congruent with a secondary moral right to forciblycoerce performance; a secondary right to forcibly coerce a monetary remedy; a secondaryprivilege to forcibly coerce performance (or a monetary remedy); a secondary right to rebuke;and other gradations as well.

49 One might further subdivide moral rights to take into account cases where pursuit of alegal remedy is morally appropriate, yet pursuit of self-help is not. Although that distinction isnot made here, it could accurately describe the moral status of contractual promises given theexistence of contract law.




B. The Moral Implications of a Private Right of Action

Even if promisees only acquire weak moral rights via ordinary, gratuitouspromises, one might still question whether the strength of these moralrights should be meaningful under an accommodationist assessment ofcontract law. Moral agents (including judges) may not perceive a relation-ship between the strength of a promisee’s moral rights—whether a prom-isee has standing to coerce when promises are breached—and the law ofcontracts. There may thus be little reason to focus on parallels betweencontract law and promissory morality in the remedial context. Would avirtuous moral agent perceive a divergence in terms of bindingness, asShiffrin does? Alternatively, would a virtuous moral agent perceive adivergence in terms of moral rights? A review of the structure and contentof the private law is helpful in analysing these questions.

Shiffrin’s divergence claim relies on a comparison of the legal significanceof breaching a contract with the moral permissibility of breaching a promise.For Shiffrin, consideration doctrine is problematic because promissorymorality requires unilateral promises to be kept, yet contract law ‘onlyregards as enforceable promises that are exchanged for something or onwhich the promisee has reasonably relied to her detriment’.50 Implicitly then,Shiffrin is interpreting the non-enforcement of contracts that lack considera-tion as a legal statement about the moral bindingness of contractual promises.Yet it would oversimplify contract law to say that courts require parties tokeep their promises when there is bargained-for consideration, and do notrequire them to keep their promises in the absence of consideration. Whatcontract law actually does is to provide for the possibility of enforcementcontingent upon an individual bringing suit. Courts do not generally enforcelegal rights sua sponte; they enforce them when a plaintiff initiates litigation.An accommodationist should thus be concerned with the legal and moraljustification for private rights of action.

Benjamin Zipursky has recently offered important insights into privaterights of action which are relevant here. Zipursky provides an illuminatingexplanation of how private rights of action, rather than corrective justiceor efficiency, are central to an explanation of the private law’s function.51

In the process, his theory also suggests how a claimant’s moral right torespond to a breach can be linked to a legal power to bring aboutenforcement.52 As will become apparent, this is an area of law where moral

50 Shiffrin, above n 1, at 710.51 See B Zipursky, ‘Private Law’ in J Coleman and S Shapiro (eds), The Oxford Handbook

of Jurisprudence & Philosophy of Law (Oxford, Oxford University Press, 2002), 623–55.52 Although this right of action is often described as a power, Zipursky has also described

it as a Hohfeldian privilege. See B Zipursky, ‘Rights, Wrongs, and Recourse in the Law ofTorts’ (1998) 51 Vanderbilt Law Review 1, 81.

128 Andrew S Gold



agents are confronted with a pervasive legal practice that involves simulta-neous participation in a moral relationship.

In Zipursky’s view, the private right of action is a key concept inexplaining the private law. The use of a private right of action means thatthe law does not simply enforce corrective justice in cases where one partyhas wronged another party. Rather than automatically imposing civilliability when such wrongs occur, the private law ‘empowers private partiesto have other private parties held liable to them, if they choose’.53 Afoundation of liability under this system is the initiation of a lawsuit; thejudicial process is optional, and dependent on private choices.

In explaining this structure, Zipursky describes two principles of theprivate law. The first principle, a right of redress, is drawn from JohnLocke. Locke famously indicates that there is a natural right to punishwrongdoers. But Locke also indicates that individuals possess a naturalright to redress wrongs committed against them. According to Locke, ‘hewho has received any damages has, besides the right of punishmentcommon to him with other men, a particular right to seek reparation fromhim that has done it’.54 This idea of a natural right to redress can readilybe found among the norms of the private law. However, Locke alsodescribes the setting up of a judge ‘with authority to determine all thecontroversies and redress the injuries that may happen to any member ofthe commonwealth’.55 Under Locke’s understanding of the social contract,the state is apparently meant to provide redress for private injuries as amatter of course. As Zipursky suggests, this aspect of Locke’s approachdoes not fit as a description of the private law.56 The state does notautomatically remedy individual injuries due to torts or breaches ofcontract. This is where William Blackstone’s contribution is important.

Blackstone emphasises the private role in determining whether redresswill occur. In explaining how the private law works, Blackstone refers tosuits in court:

wherein the act of the parties and the act of law co-operate; the act of the partiesbeing necessary to set the law in motion, and the process of the law being ingeneral the only instrument by which the parties are enabled to procure a certainand adequate redress.57

53 Zipursky, above n 51, at 624.54 J Locke, ‘The Second Treatise of Government’ in D Wootton (ed), Political Writings of

John Locke (New York, Mentor, 1993) 265–6.55 Ibid, at 305.56 Zipursky, above n 51, at 640 (noting that ‘the Lockian picture does not match the law

we actually have’).57 William Blackstone, Commentaries on the Law of England (Oxford, Clarendon Press,

1765), bk 3, ch 3, *22.




Locke argues for the right of the individual to redress of injuries, andsuggests the state’s role in providing such redress. Blackstone notes that theredress provided by the courts is one which is procured by the individualplaintiffs themselves, and consequently the judicial remedy is a redresspartly under the individual plaintiff’s control.

Zipursky contends that synthesising these two lines of thought suggests‘the view that the power to alter a defendant’s legal status through havinga judgment entered against him—the private right of action—is somethinga private party who has been wronged is entitled to from the state’.58

Moreover, this relationship is not only about an individual’s rights, it isalso about the state’s duties: ‘the state—having deprived individuals ofother means of self-help—is obligated to empower individuals with anavenue of civil recourse through the courts’.59 This is a three partynormative relationship, one existing between the wronged individual, thestate, and the wrongdoer.

From this perspective, private rights of action can be a means ofattaining corrective justice, but the normative justification for the state’sinvolvement in providing a remedy is not solely the existence of anunrectified injustice.60 The state is obligated to provide the right of actionin those circumstances where the individual bringing suit would be justlyable to seek private redress, but in a civil society lacks that recoursethrough purely private means. The claimant is still the one initiating theremedial action, but redress occurs through the means of a state-providedinstitution. In place of non-legal remedies, the claimant is able to exercise alegal power to attain redress through the state.

Importantly, this normative relationship between a plaintiff’s legal powerto initiate a lawsuit and the state’s obligation to provide recourse would besignificant to a moral agent such as Shiffrin has in mind. A moral agentwill predictably draw connections between the availability of legal recourseand a legal wrong committed by the defendant against the plaintiff. Thelegal wrong justifies the private right of action. Although Zipursky’sexplanations for the plaintiff’s entitlement to legal recourse are premisedon the existence of a legal wrong, however, the right of action also offersredress for a moral wrong. Both types of wrong would be relevant to themoral agent. The virtuous moral agent will be interested not only in thelegal justification for the existence of a right of action, but also in the

58 Zipursky, above n 51, at 642.59 Ibid.60 However, it should be noted that a right to redress is not identical to a right to

corrective justice. See J Goldberg, ‘The Constitutional Status of Tort Law: Due Process andthe Right to a Law for the Redress of Wrongs’ (2005) 115 Yale Law Journal 524, 604(suggesting that ‘redress theory is not tied to a notion of restoration or making whole’).

130 Andrew S Gold



moral question of whether the plaintiff’s pursuit of this right of action is amorally acceptable response given the moral wrong at issue.

As Zipursky suggests, ‘the explanation of the court’s affording … relief[in contract cases] is not simply the defendant’s obligation to perform; it isthe plaintiff’s right to have the defendant perform, in light of the defend-ant’s promise to her’.61 Zipursky’s focus is on wrongs that occur throughviolation of a plaintiff’s legal rights, but his insight is also applicable froma moral perspective.62 Relief is explicable in light of the plaintiff’s moralright to performance based on the defendant’s promise to the plaintiff. Theoption of bringing suit functions as an exception to the general rule thatprivate individuals cannot coerce others.63 Indirectly, the plaintiff cancoerce others through institutional means. A moral agent is likely to seethis possibility of legal coercion in terms of redress for a violation ofpromissory morality—as Shiffrin notes, under US law the contract isrepresented as an enforceable promise.64 Thus, in this setting, the moralstatus of the promisee is significant, and not merely the performanceobligation of the promisor. A promise which creates a strong moral rightlegitimises acts of coercion against the promisor (legal or non-legal), whilea weak moral right does not.

Admittedly, social contract theories are controversial. What the aboveanalysis does highlight is the likelihood that moral agents will believe thatthe private right of action is an expression of their moral rights, includingthe morality of the promisee using coercion against the promisor. Zipurskydescribes the commonly felt entitlement to a right of redress:

[I]n light of the fact that we each have instincts to redress wrongs done to us thatthe state prohibits us from acting upon, and that such a framework of rawliberty to redress wrongs would be of some value to the person who waswronged in terms of self-preservation and self-restoration, the state is obligatedto provide someone who has been wronged an avenue of civil recourse, a civilright to redress, through the courts, against the wrongdoer.65

The instincts described by Zipursky apply to both moral and legal wrongs.Private rights of action in law often overlap with rights to redress inmorality, and for virtuous moral agents, the overlap can be meaningful.

61 Zipursky, above n 51, at 646.62 See, eg, Zipursky, above n 52, at 82 (suggesting tort law can be understood as

embodying a principle of civil recourse which ‘states that a person ought to be permitted civilrecourse against one who has violated her legal rights’). See also Zipursky, above n 51, at 636(‘It is part of the concept of a private right of action as it is deployed in the law that it isalways conceived of in relation to a particular act of the defendant who is being sued, wherethat act is being characterised as to its legal status’).

63 Zipursky, above n 52, at 81; Zipursky, above n 51, at 636.64 Shiffrin, above n 1, at 721.65 Zipursky, above n 51, at 642.




The idea that certain individuals are, and should be, entitled to pursueprivate rights of action as a substitute for a moral right of self-help suggestsa morality-based structure for the private law. This is especially so in casesinvolving breaches of contract, where promissory morality is implicated bythe wrongs at issue. Private plaintiffs frequently think they should be ableto force performance of a contractual promise, using courts as a means toaccomplish this coercion. This sense of the promisee’s entitlement is notjust a product of a legal doctrine. If the wrong at issue is seen as a moralwrong, the plaintiff may understand the state-provided right of recourse asa morally legitimate means to redress that wrong, premised on theexistence of a strong moral right. Likewise, a virtuous moral agent wouldalso be concerned if the private right of action does not correspond to astrong moral right—in that case a coercive legal remedy would divergefrom morality. An accommodationist should therefore be interested inascertaining whether particular private rights of action converge withcommonly accepted moral principles, in terms of the legitimacy of forciblecoercion by the claimant. An understanding that the private right of actionis a substitute for self-help suggests that the strength of the promisee’smoral rights would be a salient issue for a virtuous moral agent, and thussuggests the importance of a parallel between the strength of a promisee’smoral rights and the availability of a private right of action.

C. The Significance of the Promisee’s Moral Rights forAccommodationism

If a moral agent would believe private rights of action for breach ofcontract should only be available when they correspond to strong moralrights, this challenges Shiffrin’s critique of the consideration doctrine.Shiffrin’s claim of divergence would work if gratuitous promises result instrong moral rights, yet it is doubtful that ordinary promises which lackconsideration achieve this result—they are often thought to create standingto demand performance or to rebuke a breach, but nothing more. Anexpansion of contractual liability to cover gratuitous promises on the basisof promissory morality would be likely to create a divergence between themoral agent’s understanding of when a promisee may legitimately attemptcoercive remedies for breach of promise, and the justification for the state’sprovision of legal remedies. The state would implicitly be recognising thatstrong moral rights stem from gratuitous promises when promissorymorality would not recognise strong moral rights. That said, the challengefor Shiffrin’s critique is potentially more significant. Depending on one’sviewpoint, coercion of the promisor in such cases is not only immoral, it is

132 Andrew S Gold



also an unjust interference with the autonomy of the promisor.66 Promisorscould have a moral right not to be forced to perform a gratuitous promise,even if performance is their ethical duty.

Shiffrin has underscored the overlap between promissory morality andcontracts. Contracts, she suggests, involve participating in a legal practicewhile simultaneously participating in a moral practice.67 It is for thisreason that accommodation of morality becomes important: legallyenforceable contracts are more than promises, yet they often implicate themorality of promising. While Shiffrin focuses on the overlap betweencontractual and promissory obligations, a similar legal and moral overlapexists with respect to remedies. Initiating a private right of action alsoinvolves participating in a legal practice while simultaneously participatingin a moral practice. The promisee who brings suit to remedy a contractviolation is, indirectly, seeking redress for a promissory violation as well.

Once the focus is on moral rights and moral remedies, the ability to useShiffrin’s transparency criterion in this setting is greatly weakened. There isbroad agreement that, in general, breaking one’s word violates a moralobligation. There is also a widespread belief that promises create standingto demand performance. On the other hand, there is no consensus insupport of the view that ordinary promises, lacking bargained-for consid-eration, create a strong moral right which thereby legitimises the promis-ee’s use of force to make a promisor perform. An absence of consensus onthis question means that efforts to comply with the transparency criterionbased on the moral views of one subset of the population could easilyviolate the transparency criterion based on the views of another subset.

The purported divergence found by Shiffrin between contract law andpromissory morality with respect to the consideration doctrine dependsupon a contested view of promissory rights, and their correspondingremedies. As noted, a leading theory of promising emphasises moral rightsto performance yet also questions whether ordinary promises legitimise theuse of force (beyond a rebuke) to coerce performance. In addition,contractual consideration can be seen as a way of acquiring strong moralrights, such that forcible coercion will become morally appropriate. In lightof the above, consideration doctrine is explicable based on normativedistinctions among promises. It precludes legal enforcement when promis-sory morality would not support coercive measures by the promisee, and itpermits legal enforcement when promissory morality arguably does sup-port coercive measures.

66 It should also be noted that moral concerns in this setting extend beyond questions ofmoral rights. As Charles Fried notes, there are moral issues that arise in the promissorycontext that are not part of the morality of promising. See Fried, above n 3, at 8 (discussingan analogy between a promisee and the Good Samaritan).

67 Shiffrin, above n 1, at 717.




In my view, there are good arguments to support the view thatbargained-for consideration supports strong moral rights. It is also com-monly believed that ordinary, gratuitous promises do not support strongmoral rights. If these positions are correct, the consideration doctrine isunproblematic from the perspective of promissory morality. However, it isnot necessary for a particular theory of contracts—or promising—tobecome the consensus view in order to challenge Shiffrin’s argument. A livedebate over whether promises for consideration can generate enforceablerights consistent with promissory morality calls into question the use ofaccommodationism as a means to cast doubt on the consideration doc-trine.

Of course, moral claims often raise indeterminacies, and occasionalnormative disputes do not challenge Shiffrin’s theory. But the appeal of heraccommodationist approach depends on a commonly accepted set of moralpremises, even if it does not require complete agreement on the sources andcontent of those moral premises. This article suggests the areas of promis-sory morality at issue for the accommodationist approach are not settled.Significant disagreements as to when a promise creates strong moral rightsrender it much harder to advocate shifts in legal doctrine based on theburdens the law’s justifications will create for virtuous moral agents. Oncethe focus shifts from a promisor’s obligation to perform—a widelyaccepted duty—to more controversial features of promissory morality,accommodationism no longer provides easy answers.

In response to these challenges, one could assume a particular moralperspective on promising is the correct one. For example, Shiffrin’sconclusions could be supported by a view of promising in which thepresence of reliance or bargained-for consideration does not affect themorality of coercive remedies for promissory breach. In that case, theimport of the accommodationist approach will vary depending on theindividual reader’s agreement with the chosen conception of promissorymorality.68 Alternatively, an accommodationist could focus only on thosemoral premises which are widely held. Yet it is questionable whether theassumptions about promising implicit in Shiffrin’s article are in fact widelyheld by persons of virtue, especially with respect to the enforcement ofordinary, gratuitous promises. Without in some way addressing theseconcerns, accommodationism does not provide an adequate justificationfor reforming the consideration doctrine.

Moreover, Shiffrin’s theory is designed to be consistent with notions thatthe right should prevail over the good.69 Her argument is intended to

68 Taking this path also runs a risk of presupposing a ‘particular comprehensive concep-tion of the good or ideal of virtue’, a possibility which Shiffrin has attempted to avoid. Ibid,at 716.

69 Ibid.

134 Andrew S Gold



facilitate moral behaviour in those areas where issues of justice underdeter-mine legal doctrine. The mailbox rule, for example, falls into this category.In contrast to the mailbox rule, the use of coercion to enforce promisesdoes raise justice concerns. If the state is understood as an indirect meansfor promisees to obtain moral remedies, the existence of the promisee’smoral right to pursue those remedies becomes important. Coercion of apromisor to perform or pay damages for breach of contract is plausiblyviewed as a violation of the promisor’s autonomy, unless there is somethingmore at issue than the mere ethical duty to keep one’s word. This concernis legitimate in cases where the promisee only possesses a weak moral rightas a result of the promise under dispute.

V. APPLYING THE CULTURAL CRITERION TO THE CONSIDERATIONDOCTRINE

Whether or not the transparency criterion justifies eliminating the consid-eration doctrine, an accommodationist critic of the consideration doctrinemight still point to Shiffrin’s third criterion—the cultural criterion—forsupport. This cultural criterion calls into question a legal doctrine if thepractical outcome of retaining the doctrine is to negatively affect themaintenance of a moral culture. Here, the focus is on the social impact of alegal doctrine, given that moral agents participate in both legal and moralcultures at the same time.

Indirectly, the consideration doctrine—it is said—can be corrosive ofimportant promissory values. Individuals may come to feel that promisesare only really binding in cases where there is a quid pro quo, forexample.70 Moral agents may notice which contracts courts choose toenforce and draw a comparison to those promises which are morallybinding. These individuals may then feel that courts do not take promisesseriously enough, and this perceived lack of judicial concern could spillover into dealings in non-legal contexts. None of these outcomes can beruled out as potential social effects of the consideration doctrine.

Assuming that promisors would in fact take their gratuitous promisesless seriously as a result of the consideration doctrine, this cultural concernlends force to an accommodationist critique of the consideration doctrine.Granting that this outcome is possible, it is uncertain what overall effectsthe law of contract has on the morality of promisors and promisees. Doesthe consideration doctrine make people feel that they don’t always have tokeep otherwise morally binding promises? This is in part an empiricalquestion. The problem is not solely empirical, however. The question of

70 Ibid, at 742, 752 n 90.




what morality requires must be answered in order to analyse whether alegal doctrine is having a negative impact on moral behaviour.

A complete assessment of the cultural issue again requires analysis of thestrength of a promisee’s moral rights, and not merely the promisor’sprimary moral obligations. Implicit in the private right of action—formany, at least—is the idea that the state is acting to permit the use ofcoercion by legal means which the promisee should otherwise be permittedthrough acts of self-help. If the doctrine of consideration were removed,the culture of promising could be altered to include the idea that coercionis an intrinsic part of a promissory relationship, available when breachesoccur. Promisees could adopt a more truculent attitude towards promisors.This outcome, if it transpired, would not be desirable.

It is ultimately unclear whether more expansive legal understandings ofpromissory rights would have a broader effect on the general practice ofmaking promises. Educated guesses are possible, but we simply don’t know.The risks to our present culture of promising could nevertheless be substan-tial. As Shiffrin suggests, the practice of promising may be fragile.71 One legalchoice may risk negatively affecting promissory duties, while another legalchoice may risk unduly expanding perceptions of a promisee’s rights. IfShiffrin is correct, the consideration doctrine risks damage to the publicunderstanding of when promises must be kept, with many individuals feelingless of an obligation to keep their word outside of quid pro quo agreements.On the other hand, the absence of consideration doctrine could increase thereadiness of promisees to resort to coercion (legal or otherwise) in order toremedy a breach, in cases where coercion would be immoral.

Shiffrin’s cultural criterion offers an important factor to consider incertain contexts. For example, it adds another concern to the debates overthe merits of efficient breach theory as a justification for legal doctrine.72

Notwithstanding this potential relevance to contract law, the effects of theconsideration doctrine on a moral culture of promising remain obscure.The empirical question here is not only uncertain, it may actually beunanswerable, even if the attendant moral questions are decisivelyresolved.73 Absent stronger reason to suspect the consideration doctrine isdamaging to a moral culture of promising, this cultural criterion does notsupport revisions to the consideration doctrine.

71 Ibid, at 738: expressing concern that ‘the culture of trust and promising is fragile insubtle ways that are difficult to track’.

72 Ibid, at 742 (raising the possibility that the law, through its structure or justifications,could encourage individuals to ‘engage freely in promissory breach when breach yields onlymarginal economic gains’).

73 In social science terms, this appears to be a ‘trans-scientific’ problem. See A Vermeule,Judging Under Uncertainty 158 (Cambridge, Harvard University Press, 2006) (describinghow a question may be ‘trans-scientific’ when it is an empirical question, but ‘also (in manyinstances) unresolvable at acceptable cost within any reasonable time frame’.)

136 Andrew S Gold



VI. CONCLUSION

An accommodationist stance need not require us to jettison the considera-tion doctrine. An accommodationist stance does suggest the significance ofdetermining when a promisee’s moral rights cross the line between a rightto demand performance, and a right to force performance or its financialequivalent. Shiffrin’s argument works as well as it does because it focuseson an area of promissory morality which is relatively uncontroversial. It iswidely accepted that promises create moral duties for promisors to keeptheir word, and a divergence between law and morality on this issue wouldnaturally be problematic for a virtuous moral agent. Once the focus shiftsto the moral remedies available to promisees, however, this apparentconsensus proves evanescent.

Accommodationists should not conflate a promisee’s right to demandsomething with a promisee’s moral right to coerce the thing demanded.There is a continuum of moral rights at issue, and the effect of a particularfact pattern on a promisee’s moral rights should be legally relevant. Theconsideration doctrine is plausibly related to the morality of a promisee’sefforts to obtain coercive remedies under particular factual circumstances.In addition, the legal justifications for a private right of action suggest thata virtuous moral agent would draw a parallel between the availability oflegal remedies and the morality of using coercion to remedy a promissorybreach. Given these possibilities, the consideration doctrine could readilybe consistent with promissory morality.

If a convergence between morality and contract law is sought, thesemoral questions merit further inquiry. Yet there is no answer under anaccommodationist approach as to precisely when promissory moralityparallels contract enforcement, unless a particular view of promissorymorality is adopted. It is possible to imagine legal doctrines which wouldbe beyond the pale for a virtuous moral agent under almost any conceptionof the morality of promising. However, such cases are not at stake here.Under reasonable understandings of promissory morality, the considera-tion doctrine is appropriate. Absent a significant shift in the understand-ings of virtuous moral agents, the consideration doctrine should thusremain an acceptable feature of contract law, even for an avowed accom-modationist.






6

Justifying Damages

CHARLIE WEBB

WHAT CLAIMS SHOULD be available to a claimant who hassuffered a breach of contract? The aim of this article is toexamine how we may justify particular awards and orders that a

court may make in such cases. The focus will be on monetary awards—ordamages1—though the analysis which follows also bears on other possibleorders.

Now, the possible aims of damages awards are at least as broad as thepossible aims of contract law generally. In the first instance, our under-standing of the incidence and form of damages awards following breachesof contractual obligations will derive from our view of the basis and scopeof those obligations (whether or not such awards aim at replicating orenforcing such obligations). Moreover, any argument as to the justifiabilityof particular damages awards, as with any argument as to the propercontent of some part of private law, ultimately rests on a particularconception of the legitimate functions and goals of private law. Though afull account and justification of damages awards must, therefore, depend

1 The proper scope of the term ‘damages’ is controversial. On one view, we should reservethis terminology for loss-based monetary awards. The modern tendency, however, is to extendthe language of ‘damages’ to all monetary awards for wrongs, whatever their basis ormeasure: see, eg, J Edelman, Gain-Based Damages: Contract, Tort, Equity and IntellectualProperty (Oxford, Hart Publishing, 2002) 5. The issue is particularly confused in relation tocontract law, for, while many seem happy to apply the label ‘damages’ to a variety of awardswhich serve clearly different purposes and have clearly different bases (eg ‘expectation’damages, reliance damages, restitutionary damages), I have not seen anyone refer to claimsfor payment of an agreed sum (or debt) as claims for damages. This makes sense on the basisthat the obligation enforced through such actions is the defendant’s primary obligation to pay,an obligation which arises at the point of contract formation rather than later when thecontract is breached. As such, these are not monetary awards given for breaches of contract.However, contract lawyers do tend to attach the label ‘damages’ to other awards whichsimilarly seek to effectuate the claimant’s primary right to performance: see, eg, the discussionof ‘substitutive damages’ in the text accompanying nn 58–64. For present purposes, I shall usethe term ‘damages’ to embrace all monetary awards available where a claim is broughtfollowing a breach of contract, irrespective of whether the award is, strictly speaking, for thewrong of breach.



and be built upon a particular theory of contractual obligation and ofprivate law more generally, I shall not address these questions in any detail.Instead, I shall be focusing on a narrower question: namely, what are theconsequences of the recognition of a right to performance when consider-ing how the law should deal with claims brought following breaches ofcontract? What sort of awards can such a right support?

I should make one point clear at the outset. The approach andarguments developed here are not offered as an account or elucidation ofthe existing rules on contract damages. Some parts are consistent with thelaw as it stands, others are not. Rather, the aim of this article is to explorewhere acceptance of a right to performance takes us when addressing thequestion of how the law should respond to breaches of contract. In so faras the law deviates from the position taken here, a number of conclusionsare possible. One is that the law, though committed to the view thatcontracts entail rights to performance, also recognises other principleswhich rightly limit the protection offered to such rights. Another is that,upon closer examination, the law does not, and is right not to, recogniserights to performance as a consequence (or at least as an inevitableconsequence) of contract formation. A third possibility is that the law isstraightforwardly defective in failing to give adequate recognition to theright to performance when dealing with claims for breach of contract. Fornow, the key point to note is that the arguments made here are notpremised upon the correctness of, and so they do not stand or fall withtheir ability to fit, the cases.

I. IDENTIFYING THE RIGHT TO PERFORMANCE

That contracts are a source of legal rights and obligations is not in doubt.What is less clear is to what rights and obligations contracts give rise.Those who claim that the law does or should recognise contracting partiesas having an interest in, and right to, performance argue that a typicalbilateral contract consists of a set of reciprocal undertakings and that acontracting party, upon entering into a contract, acquires a right that hercounterpart fulfil or make good her undertakings, while coming under aduty to fulfil her own. So, if I contract to buy your shoes, I acquire a rightthat you transfer the shoes to me and come under an obligation to pay youthe purchase price. You in turn acquire a right that I pay you the agreedprice and become subject to an obligation to transfer the shoes to me. Thisis not uncontroversial. It is not self-evident that entry into a contractshould give a claimant a right to the defendant’s performance, as opposedto, for instance, a right to the monetary equivalent of performance or aright simply to have compensated losses caused by a failure to perform.Oliver Wendell Holmes famously argued that ‘[t]he duty to keep a contract

140 Charlie Webb



at common law means a prediction that you must pay damages if you donot keep it,—and nothing else’.2 The existence of a right to performancecan and should be questioned.

The scope and content of individual rights and obligations follow fromand are determined by the principles that underlie them, and, as such, a fulldefence of the right to performance would involve an inquiry into why weconsider that legal obligations, of any sort, should arise from agreementsor mutual undertakings. This would require, at the very least, an article ofits own. For present purposes it is enough to point out that a rejection ofthe right to performance also requires us to reject the notion of breach ofcontract, at least as conventionally understood.3 A breach of contract is afailure of the defendant to perform the contract according to its terms. Butwithout a duty to perform, there can be no breach in the defendant’ssimple failure to adhere to the terms of the contract. A claimant’s right toperformance is no more than the correlative of the defendant’s duty toperform. So, unless we are to say that, whenever we talk of breach ofcontract, we are really engaged in a sort of fiction or deception, we mustaccept that there is, in some form, a right to performance.4 Accordingly,commitment to the notion of breach of contract necessarily involvescommitment to the notion of a right to performance.5

In any case, I shall proceed on the assumption—shared, I think, by themajority of contract lawyers—that such a right does exist. For those whodoubt whether there really is or should be a right to contractual perform-ance, they may understand the argument as taking the form ‘if it is true tosay the we have a right to performance, then …’. The validity of thepremise must be established, if at all, another time.

2 OW Holmes, ‘The Path of the Law’ (1897) 10 Harvard Law Review 457, 462. See, tosimilar effect, OW Holmes, The Common Law (Boston, Little, Brown and Co, 1881) 301:‘the only universal consequence of a legally binding promise is, that the law makes thepromisor pay damages if the promised event does not come to pass. In every case it leaves himfree to break his contract if he chooses.’ See too, D Campbell and D Harris, ‘In Defence ofBreach: A Critique of Restitution and the Performance Interest’ (2002) 22 Legal Studies 208.

3 I make a fuller version of this argument in C Webb, ‘Performance and Compensation:An Analysis of Contract Damages and Contractual Obligation’ (2006) 26 OJLS 41, 46–7.

4 For instance, Holmes’ argument requires not only that we deny the existence of anyobligation to perform, and correlative right to performance, but also that we reject the viewthat claims for compensatory damages are founded on the commission of a breach ofcontract. On this approach, there could be no such thing as a claim for breach of contract butonly a claim that the defendant fulfil his duty to pay damages in the event of non-performance. See too, Webb, ibid, at 45–9.

5 This leaves unanswered the important question of what performance entails and howwe go about determining this: for instance, is the content of the obligation determined byreference to the undertakings the defendant intended to make or to what the claimant(reasonably) understood the defendant to be undertaking?

Justifying Damages 141



II. RESPONDING TO BREACHES OF CONTRACT

As we have seen, a breach of contract is committed where a defendant failsto fulfil his contractual undertakings, and hence where the claimant doesnot receive the performance to which he or she is entitled. A breach ofcontract may or may not leave a claimant worse off. It may or may notbring a gain to the defendant. A breach of contract which causes theclaimant no loss is a breach of contract nonetheless.

So, if in our earlier example you fail to deliver the shoes, you commit abreach of contract. We do not first need to inquire into the value I place onthe shoes, and hence on your performance, in order to determine that yourfailure to deliver them amounts to a breach. So, if I have now decided thatI do not like the shoes or have found a better pair elsewhere, while thismay well have a bearing on my decision to sue and on what a court mightthen award me, it does not make your failure to deliver any less a breach.This is just a roundabout way of showing that my right to performance ismore than, and so is not reducible to, a right not be left worse off in theevent of your failure to perform. I have a right that you performirrespective of what losses, if any, I may suffer in the event that you fail todo so.6

This may be elementary, but it is important not to lose sight of this whenwe ask how the law should deal with claims brought following breaches ofcontract. Breaches of contract will often cause the claimant a loss. In suchcases, this loss does itself ground a claim and provides a reason for anaward.7 However, as we have seen, the claimant’s rights do not stop there.The claimant has a right to performance, and this right is not satisfiedsimply by ensuring that she is not left worse off as a result of thedefendant’s failure to perform. This right can be asserted, and should besufficient to ground a claim, whenever a contract is breached, irrespectiveof what losses that breach may have caused and independently of anyclaim that may exist to have those losses made good.

Of course, the most obvious way of giving effect to the claimant’s rightto performance is by ordering the contract to be specifically performed.This is what we do as a matter of course with obligations to pay money,which are enforced through the action for an agreed sum or debt.However, outside such obligations, the courts order specific performanceexceptionally, only where an award of damages is held to be inadequate.

6 Of course, none of this is to suggest that the law should not also recognise the claimantas having a right to have compensated losses caused by the defendant’s failure to perform. Itshould and it does. The point is that the right to performance exists independently of, andcannot be reduced to, any right to avoid or have made good such losses: see further thediscussion of Holmes’ theory, above n 2 and accompanying text. See too, Webb, above n 3, at45–9.

7 See text accompanying nn 42–57.

142 Charlie Webb



The conventional view is that damages awards are intended to compen-sate the claimant for the losses she has suffered as a result of not receivingthe performance to which she is entitled.8 As such, though there remainssome uncertainty as to what losses fall to be compensated and how theyare to be assessed, the claimant is entitled only to such damages as arenecessary to make good any relevant losses. As a corollary of this, wherethe breach has caused no loss, the claimant can recover no more thannominal damages.9 However, the rule that specific performance will beavailable only where such an award is considered inadequate poses anumber of problems.10

There are undoubtedly good reasons for preferring the simplicity andfinality of monetary awards, and so we should be happy to see the courtsprioritising damages awards over specific relief where the two providealternative and otherwise equally effective routes to achieving the same endresult. However, if we accept the conventional view of damages awards,specific performance and damages can be seen to serve different purposes.While specific performance is designed to secure for claimants the perform-ance for which they contracted and to which they are entitled, damagesseek not to provide claimants with that performance but instead to ensurethat they are not left worse off as a result of not receiving that perform-ance. To obtain performance is one thing; to receive a sum of money tomake up for the losses caused by not obtaining performance is somethingdifferent.

8 See Robinson v Harman (1848) 1 Exch 850, 154 ER 363, 365 (Parke B); RuxleyElectronics and Construction Ltd v Forsyth [1996] AC 344 (HL) 357 (Lord Jauncey), 360(Lord Mustill); Golden Strait Corpn v Nippon Yusen Kubishika Kaisha (The Golden Victory)[2007] 2 AC 353 (HL) [29] (Lord Scott).

9 I put to one side claims to gains made by the defendant through breach which mayentitle claimants to recover a sum of money in excess of any losses they have suffered.

10 These problems extend beyond those detailed in the text. For instance, to apply theadequacy of damages test, we must first know what the goal of a damages award is, sinceonly then can we determine whether, on a given set of facts, such an award would do that jobadequately. On the orthodox view that damages are compensatory, and given that all lossesare necessarily capable of being made good by a monetary payment (see below nn 49–57 andaccompanying text), it follows that damages will be inadequate only where the extent of theclaimant’s losses is uncertain and hence where we do not know what sum to award. However,this does not tally with the cases in which specific performance tends to be ordered; forinstance, there is no reason to think that valuing land is any more difficult than valuing otherassets, and yet contracts for the sale of land are routinely specifically enforced. See too SSmith, ‘Substitutionary Damages’ in C Rickett (ed), Justifying Private Law Remedies (Oxford,Hart Publishing, 2008) 93, at 105–9. The likely explanation is that courts have beenmotivated to order specific performance where compensatory damages have been viewed asinadequate for the purpose of securing the performance to which the claimant is entitled.However, as argued in the text, securing performance and receiving compensation for thelosses caused by non-performance are different things, and, therefore, damages aimed atcompensating the claimant for such losses do not seek to secure performance and so will beadequate for this latter purpose only occasionally and never by design. See further, Webb,above n 3, at 51–3.




The adequacy of damages test, therefore, suggests that, when faced witha claim for breach of contract,11 the default response is not to giveclaimants the performance they are due but only to make good the lossescaused by non-performance. As such, at the end of the day, the claimant isleft without that performance but with something else of equal value. Thiscreates an apparent incongruity. At the point of contract formation andwhen identifying breaches of contract, we regard the claimant as having aright to performance—and the defendant as owing a duty to perform—independently and irrespective of what losses, if any, may result from afailure to perform. However, when it comes to remedying breaches ofcontract, we appear to disregard the claimant’s right to performance andinstead are concerned with ensuring only that the claimant is not left worseoff as a consequence of not receiving performance.

It looks like something has to give. If the claimant has a right toperformance then, where possible, we should see that she obtains thatperformance. If we consider that the undesirability or impracticability ofordering specific performance is sufficient reason to limit the claimant tocompensatory damages, then this calls into question the reality of theclaimant’s ‘right’ to performance.12 Such difficulties are, however, avoidedif we can find a way of giving effect to the claimant’s right to performancethrough an award of damages, since we can then have the advantages of amonetary award without denying the claimant the performance to whichshe is entitled.

III. PERFORMANCE INTEREST DAMAGES

I have argued previously that the right to performance may be capable ofbeing given effect by, and hence may justify, an award of damages assessedby reference to the sum of money it would take for the claimant topurchase from some alternative source the performance the defendant wassupposed to provide; in other words, damages assessed on a cost of cure

11 Or, more precisely, when faced with a breach of a contractual obligation other than topay a sum of money. As we have seen, where the obligation breached is to pay money, theobligation is routinely specifically enforced through the action for an agreed sum.

12 The question here is whether we can say that the claimant has a legal right toperformance, and the defendant a legal duty to perform, if the law will in no circumstancesrequire the defendant to perform even when in a position to do so. It is difficult to see whyconsiderations, such as a concern with unduly interfering with the defendant’s liberty and thecost and complexity of supervision, which bear on the appropriateness of compellingperformance when the courts are faced with a claim for breach of contract, should not equallybe factored in when determining whether a right to performance arises in the first place.Either such considerations outweigh the arguments for requiring the defendant to perform orthey do not. If they do, we should conclude that the claimant has no right to performance; ifthey do not, the claimant should have such a right and the courts should give effect to it. Seetoo P Jaffey, Private Law and Property Claims (Oxford, Hart Publishing, 2007) 39–45.

144 Charlie Webb



basis.13 I shall only sketch the argument here and address some of the keyfeatures of, and potential challenges to, such an award.

As we have noted already, upon entry into a contract the claimantacquires a right to its performance irrespective of the value she places onthat performance and the losses she would suffer in the event of non-performance. Traditional compensatory damages awards accordingly failto give effect to the claimant’s right to performance as their purpose is onlyto ensure that the claimant is not left worse off as a result of breach, ratherthan to see that the claimant obtains the very performance for which shecontracted and which she is due. An award of damages will give effect tothe claimant’s right to performance only where and to the extent that theaward provides the claimant with that performance. However, in so far assuch a result can be achieved through an award of damages, we can justifysuch an award by reference to the claimant’s right to performance. Thequestion then is how and in what circumstances a damages award can besaid to provide the claimant with performance.

Where the claimant has contracted for some asset or service, it is thedefendant’s duty to provide the claimant with that asset or service. Wherethe defendant fails to do this and the court is unwilling to compel thedefendant to provide that asset or service herself, it may still be possible forthe claimant to purchase an equivalent asset or service, and hence anequivalent performance, from a third party. The claimant’s right toperformance can, therefore, be given effect, and the defendant’s duty toprovide that performance enforced, by requiring the defendant to pay theclaimant the money needed to obtain an equivalent asset or service from athird party. Albeit by a different route, the claimant ends up with theperformance for which she contracted and to which she is entitled.

There are, however, a couple of provisos. Firstly, the claimant’s right is toperformance and the award of cost of cure damages is premised on the factthat the sum will be used to obtain that performance. Only if the money isso used can it be said that the award gives the claimant the performancefor which she contracted. Accordingly, a cost of cure award which is notused to secure performance cannot be justified on the basis that iteffectuates the claimant’s right to performance. In principle, therefore, suchawards should be conditional on the claimant actually using that money toobtain the relevant asset or service from another source.14

13 Webb, above n 3. Similar arguments are to be found in B Coote, ‘Contract Damages,Ruxley, and the Performance Interest’ [1997] CLJ 537; E McKendrick, ‘Breach of Contractand the Meaning of Loss’ (1999) 52 Current Legal Problems 37; M Eisenberg, ‘Actual andVirtual Specific Performance, the Theory of Efficient Breach and the Indifference Principle inContract Law’ (2005) 93 California Law Review 975; and, in particular, Smith, above n 10.

14 Webb, above n 3, at 62–3; Smith, above n 10, at 103. This, of course, contradicts theorthodox principle that what claimants do with any money awarded to them as damages isres inter alios acta. This principle makes good sense in relation to compensatory awards,




Secondly, for an award of damages to be justified on the basis that iteffectuates the claimant’s right to performance, the alternative or substitute‘performance’ which the claimant is to purchase from a third party musteither exactly replicate the performance to which the claimant is entitledunder the contract or differ from that performance only in ways which wecan regard as immaterial and so which can be disregarded. So, if I contractwith you to do some building work for me and you breach, a cost of cureaward can be justified on the basis that it gives me the performance Icontracted for only if paying another builder to come in to complete thework can be said to equate (or, at least, be tantamount) to performance ofour contract.

Sometimes there will be no difficulty in coming to this conclusion.Wherever a contract allows for the relevant work to be subcontracted out,performance ‘through’ a third party is clearly possible. However, difficul-ties arise where the contract provides, and my right is, that you do thework. In such cases the performance to which the claimant is entitledcomprises not simply an end product but embraces the means by whichthat end product is provided. How then can we say that the provision ofthat end product from an alternative source equates to performance?

I have argued previously that there are two bases which could enable usto say that, in such cases, a cost of cure award provides the performancethe claimant contracted for.15 The first would be to argue that the contractdoes in fact provide for the defendant paying a third party to providereplacement ‘performance’ in the event of the defendant herself beingunable or unwilling to perform. In most cases this could only be on thebasis of the existence of an implied term to that effect, and then the keyquestion is as to the legitimacy of implying such a term. The second wouldbe to say that, though paying a third party to provide the service or assetwhich the defendant had undertaken to provide herself cannot be said toamount to literal enforcement of the contract (and hence differs from theperformance to which the claimant is entitled), it is nonetheless closeenough to the performance required under the contract that such an awardcan be regarded as effectuating the claimant’s right to that performance.The issue then is how close is close enough. Certainly some deviation fromthe exact performance set down in the contract is permitted under theexisting rules on the availability of specific relief, where, at the very least,the ‘performance’ ordered by the court will almost invariably take place at

where the mere receipt of the money makes good the claimant’s loss and hence fulfils thepurpose of the award (see text accompanying nn 49–57). However, it cannot square withdamages awarded on the basis that they effectuate the claimant’s right to performance, sincehere the money does not in itself provide the claimant with performance but only a means ofobtaining it. As such, and in contrast to payments made as compensation, we should beconcerned with how that money is then used.

15 Webb, above n 3, at 58–61.

146 Charlie Webb



a later date than originally stipulated.16 Equally clearly, however, theremust be a limit to the extent to which we can deviate from or qualify theterms of the contract if we are to be able to justify a damages award on thebasis that it provides the claimant with the performance for which shecontracted and to which she is entitled.17

The argument here marks out a different role for cost of cure awardsthan the conventional view which sees them as one possible measure ofcompensatory damages. Indeed, where the claimant has incurred or willincur the expense of remedying the defendant’s defective performance orobtaining that performance from a different source, such a sum ispotentially recoverable as a loss consequential on the defendant’s breach ofcontract. However, the law places limits on the losses which may berecovered in a compensatory damages claim, and, in particular, it requiresthat the claimant take reasonable steps to mitigate her losses. This raisesproblems where the cost of securing performance exceeds the value theclaimant puts on that performance.18 The argument here is that protection

16 See G Treitel, Remedies for Breach of Contract (Oxford, Clarendon Press, 1988) 1.17 Smith, above n 10, suggests that the argument for awarding cost of cure damages as a

means of securing for claimants the performance to which they are entitled can be extended toclaims brought following the commission of torts. Certainly the argument that a damagesaward can be justified where it gives effect to a right held by the claimant can apply whateverthe source of that right and so is capable of extending beyond contractual rights and claims.However, I am doubtful as to how far the primary rights protected through the law of tortscan be effectuated in this way. As we have seen, where the performance to which the claimantis entitled is the provision of some service or asset, it will often be possible to acquire anequivalent service or asset, and hence an equivalent performance, via a third party. Theprimary rights and duties which underlie the law of torts tend to take a different form,however, requiring instead that the defendant refrain from acting in a particular way.Accordingly, though further breaches can be prevented by injunction, in relation to breacheswhich have already been committed it will usually be too late to secure performance. So, onceI have been assaulted or defamed, it is necessarily too late to give effect to my right that younot assault or defame me on that occasion. History cannot be rewritten and in no sense canan award of damages, however assessed, be viewed as giving me the performance—that younot assault or defame me—to which I was entitled. All that I can require is that you makegood the losses you have caused by the assault or defamation and that you not assault ordefame me in the future. Because of this, I suspect that cost of cure awards can be justified onthe basis identified in the text only where we can formulate the defendant’s primary duty as aduty to secure a certain advantage or end result (even if this is no more than the maintenanceof the status quo) for the claimant. This cannot be done in the assault and defamationexamples, but it may be possible in relation to, for instance, negligent property damage.Though typically we say that the duty here is to take reasonable care not to damage ordestroy another’s property, it could easily be reframed as a duty to take reasonable care toensure that other’s property remains undamaged by our actions. This could be broken downinto a duty, in the first instance, to take reasonable care to refrain from damaging others’property and, secondly, when it is threatened or damaged by our actions, to take positivesteps to (re)secure it (cf R v Miller [1983] 2 AC 161 (HL)). So conceived, we can see how itmight be possible to say that a cost of cure award enforces that duty and effectuates theclaimant’s correlative primary right.

18 See Smith, above n 10, at 110.




of the claimant’s performance interest may provide an alternative justifica-tion for a cost of cure damages award and which, importantly, is notsubject to the same limits as compensatory claims.

The mitigation rule denies the claimant recovery in respect of losseswhich, though (but for) caused by the defendant, the claimant couldreasonably have avoided. In effect, we are saying that it is the claimantrather than the defendant who is primarily responsible for those losses. Onthis basis, we conclude that they should be borne by the claimant ratherthan by the defendant.19 The fact that we have such a rule should not bethought surprising. Compensation claims are necessarily concerned withthe question of who should bear certain losses, and it makes good sense forthe parties’ respective contributions to such losses to be a considerationthat the courts take into account when determining by whom those lossesshould be borne.20

Different considerations, however, apply where the claimant is seeking toenforce her right to performance. The claimant is not asking for losses shehas suffered to be shifted onto the defendant; the claimant is simply askingfor the contract to be performed. If it is objected that a performanceinterest award enables the claimant to circumvent the mitigation rule andallows her to recover—and requires the defendant to pay—a sum of moneyin excess of her compensable loss, the answer is that the claimant’s rightsare not limited to recovery of losses caused by non-performance.21 Theclaimant has a right to performance itself, regardless of what value sheplaces on that performance and irrespective of what losses, if any, she maysuffer in the event of breach.22 If it is argued that such an award leaves theclaimant with a windfall, then the answer is that this is the very windfallfor which the claimant contracted and which the defendant undertook toprovide.23

19 Or, at least, we do not have sufficient reason actively to shift the loss onto thedefendant, and so allow it to lie where it falls.

20 Similarly there is good reason for taking into account the claimant’s contributorynegligence when dealing with compensatory claims (though not in relation to claims whichamount to assertions of the claimant’s performance interest); cf ForsikringsaktieselskapetVesta v Butcher [1989] AC 852 (HL); Barclays Bank Plc v Fairclough Building Ltd [1995]QB 214 (CA).

21 Cf Ruxley Electronics and Construction Ltd v Forsyth, above n 8, at 367, 369 (LordLloyd). The same objection can be made against, and similarly fails to defeat, claims forspecific relief: see, eg, White & Carter (Councils) Ltd v McGregor [1962] AC 413 (HL);Beswick v Beswick [1968] AC 58 (HL).

22 See text accompanying nn 2–6. Nor is the claimant’s right to performance conditionalon his placing upon it a value which is greater than, or at least proportionate to, the cost tothe defendant of providing it.

23 Cf Ruxley Electronics and Construction Ltd v Forsyth, above n 8, at 357–8 (LordJauncey). Of course, we are still proceeding on the basis that the claimant is required to useher damages to secure performance from some other source. Without this restriction on theuse of the money the windfall objection would often be valid.

148 Charlie Webb



In short, if we accept that the claimant has a right to performance, thenshe should be entitled to a cost of cure award, not because this is themeasure of the claimant’s compensable loss, but because this gives theclaimant the performance to which she is entitled.

IV. THE OUTER LIMITS OF PERFORMANCE

Some obligations once breached cannot then be performed. Goods deliv-ered late cannot then be delivered on time; the confidentiality of informa-tion wrongfully disseminated cannot subsequently be restored andpreserved. Inevitably the claimant cannot then demand performance ofthose obligations. At this point it appears that the argument for a damagesaward which gives effect to the claimant’s performance interest runs out,and that, whatever justification there may be for an award of substantialdamages, it cannot lie in the protection and effectuation of the claimant’sright to performance.24 In such cases,25 courts typically award the claimantcompensatory damages—that is, damages aiming to make good the losscaused to the claimant by virtue of not receiving the performance to whichshe is entitled. I shall turn shortly to look at the basis and focus of suchawards. Before then, however, I should make clear why we must lookbeyond the claimant’s right to performance if we are to justify suchawards.

My argument thus far has been that the claimant’s right to performancecan support a damages award where and to the extent that this providesthe claimant with the performance to which she is entitled. Necessarily,therefore, where performance is no longer possible, there is no scope forawarding damages which seek to secure that performance and which arejustified on that basis. Of course, claimants for whom breach has renderedperformance unachievable can still complain that their rights have beeninfringed. Moreover, we can say that any award then made, at least in sofar as it seeks to respond to or to correct the consequences of thatinfringement, is premised upon or grounded in the claimant having a rightto performance. However, we need to be able to say why a particularawarded is a justifiable response to that infringement—in other words,why the claimant should be awarded a particular sum rather than another,why any sum and not, say, a declaration. In situations where performance

24 The one qualification is that there may be other obligations under the contract whichare still capable of being performed, and so the claimant may still be able to demandperformance (whether specifically or through a damages award) of those obligations.

25 And indeed, as the law stands, in many where performance remains possible.




is impossible, and so where an award does not and cannot provideperformance, the claimant’s clear right to that performance cannot providean answer.26

Some would argue that this is all a little premature, and that the right toperformance can in fact support a damages award even where thatperformance is now beyond reach. One such argument is founded on thefact that the parties may themselves have provided in the contract for whatis to happen in the event of breach. If they have agreed that, upon breach,the defendant is to pay the claimant a particular sum, whether fixed inadvance or determined in light of the actual consequences of breach, thenan order that the defendant pay that sum to the claimant can be justifiedon the same basis as the claims we have been examining until now. Theaward ensures that the claimant receives the sum to which she is, in thecircumstances, entitled. Accordingly, though such a claim resembles aconventional award of compensatory damages, it is in fact an example of aperformance interest claim, mirroring an action for an agreed sum.

So much is clear. However, on one view contracts always make provisionfor what the parties’ respective rights and duties are in the event of breach.If this is true, the availability and scope of all claims following a breach ofcontract can be said to derive from the terms of that contract and hencecan be supported by reference to the claimant’s entitlement to the perform-ance of that contract. Of course, not all contracts provide expressly forwhat is to happen should one of the parties breach. Nonetheless, thesuggestion is that, even where the express terms are silent on this question,an answer can always be found by identifying a relevant implied term.

What should we make of this? If the claim is that the parties’ respectiverights and obligations in the event of breach are always to be located in theparties’ agreement, in the explicit or implicit undertakings they made toone another upon entry into the contract,27 I find it implausible. Contract-ing parties will generally, perhaps invariably, be aware that a failure toperform constitutes a legal wrong which can form the basis of a legalclaim, and that the court will respond to such a claim by requiring theparties in breach, in some way, to make good their wrong, (usually)through the payment of damages. But I think this still leaves us some way

26 It should be clear that I am not contending that the question of what rights we haveagainst others should be decoupled from the question of how courts should respond toinfringements of these rights. Far from it. Where the factual advantages which the claimant’sright seeks to secure can still be achieved, the court’s principal response should be to ensurethat this happens.

27 See, eg, A Kramer, ‘An Agreement-Centred Approach to Remoteness and ContractDamages’ in N Cohen and E McKendrick (eds), Comparative Remedies for Breach ofContract (Oxford, Hart Publishing, 2005) 249; C Czarnikow Ltd v Koufos (The Heron II)[1966] 2 QB 695 (CA) 730–31 (Diplock LJ); and, most recently, the speeches of LordsHoffmann and Hope in Transfield Shipping Inc v Mercator Shipping Inc (The Achilleas)[2008] UKHL 48.

150 Charlie Webb



short of saying that we can always find genuine, though tacit, undertakingscovering the parties’ responsibilities after breach.28

Firstly, awareness that our actions can or will attract legal consequencesis not limited to contracting contexts and contractual obligations. I knowthat punching you on the nose will have various legal repercussions, but werightly do not view the liability that attaches to me when I do so asderiving from some (implicit) undertaking I have made to that effect. Andthat is not (just) because of a lack of consideration or intent to create legalrelations but, more fundamentally, because there is no undertaking. Now,of course, where I walk up and punch you there are no undertakings of anysort, whereas, as between contracting parties, there will be any number ofundertakings covering other aspects of their dealings with one another. Assuch, the search for an undertaking in respect of the consequences ofbreach makes more sense and is more likely to bear fruit. Nonetheless, Ithink the point stands: knowing—indeed accepting—certain legal conse-quences will follow from our actions is not the same as undertaking to dowhatever it is the law requires of us.29

Secondly, even if we were to say that knowledge and acceptance ofcontractual rules dealing with the consequences of breach were sufficient

28 Kramer, ibid, at 256–7, appears happy to accept that contracting parties may not intendor undertake to pay damages in the event of breach. Nonetheless, he argues that such awardsare grounded in the parties’ agreement on the basis that all (primary) obligations are ‘orientedto particular consequences’ and so the ‘risks of such consequences are allocated within thecontract’. The implication here is that any given contractual undertaking (and hence anyprimary contractual obligation) is oriented towards securing for the claimant certain factualadvantages, and that these may extend beyond the specific factual advantages which form theexpress content of that undertaking (since otherwise inquiring into an obligation’s ‘orienta-tion’ would not, as Kramer intends, support liability in respect of consequential losses orindeed justify any award where performance of that explicit undertaking is no longerpossible). This is no doubt true. Many contracts are made to achieve what may be calledulterior objectives or benefits, that is, benefits beyond, though dependent upon, the simpleprovision of the promised performance. (For instance, I may want a new crankshaft for mymill so that it can continue to operate and so secure for me the trading profits which wouldfollow from its operation.) The question is where this takes us. Even where the defendantknows what ulterior objectives or benefits the claimant has in mind when contracting, it doesnot follow that we can say that the defendant undertook (or that the claimant can reasonablyunderstand the defendant to have undertaken) to secure such benefits or to compensate theclaimant for the losses she may suffer if those benefits are not secured. And, if we cannotidentify any such undertaking then we are left looking for some other justification for such anaward. Furthermore, often the parties will, reasonably, have different ulterior objectives andbenefits in mind. In such cases, even looking at the parties’ conduct objectively, there will beno single correct answer to the obligation’s orientation. So, I doubt whether an inquiry intoan obligation’s ‘orientation’ will often, let alone always, yield a single answer, and, even wherewe can identify the ulterior benefits the contract is oriented to securing, we need a reason forsaying that the defendant should be liable to compensate the claimant for the losses shesuffers from being deprived of those benefits.

29 And similarly, the fact that you know that I am aware of the legal consequences of myactions does not mean that you believe—and does not in itself provide reasonable grounds forbelief—that I am undertaking to do what the law requires me to do. Cf The Heron II, aboven 27, at 730–31 (Diplock LJ).




grounds to infer undertakings to pay damages in the event of breach, Isuspect that this would take us only so far. While most contracting partieswill know that the law provides for damages to be paid in the event ofbreach, far fewer are likely to be familiar with the detail of these rules. Insuch circumstances, the most we can infer is an undertaking to pay suchdamages as the law requires to be paid. And, if this is true, then looking tothe parties’ undertakings cannot provide any concrete guidance as to whatdamages should be awarded, nor can they support any particular measureover others. Once more, we must look beyond the claimant’s right toperformance to justify the award.

So, inquiring into the parties’ agreement or undertakings will sometimesyield an answer to what their respective rights and obligations are in the eventof breach, but all too often it will not.30 Accordingly, the claim that theconsequences of breach—and hence the existence and scope of any damagesaward—can always be attributed to a relevant provision in the contract canonly hold true if our inquiry into the terms of the contract does not stop at theparties’ express and implied undertakings. Of course, we do use the languageof implied terms to cover obligations owed by contracting parties which donot derive from their agreement and the undertakings they have given to oneanother.31 And, though we typically limit this language to primary terms ofthe contract, there is no reason why it cannot or should not be extended toobligations imposed on the parties following breach. Indeed, as we have seen,parties are free to provide expressly for their rights and obligations uponbreach, and so we could say that, where no such provision has been made,there is necessarily a gap in the contract which can only be filled by theimplication of a relevant term.

30 Accordingly I agree with the basic approach of English courts in cases where the basisor measure of damages is at stake, by which the answer is not thought to be found inexamination of the terms of the parties’ agreement: see, eg, Ruxley Electronics andConstruction Ltd v Forsyth, above n 8; Attorney-General v Blake [2001] 1 AC 268 (HL);Alfred McAlpine Construction Ltd v Panatown Ltd [2001] 1 AC 518 (HL); cf the speeches ofLords Hoffmann and Hope in The Achilleas, above n 27. See too the statement of Bowen LJin Birmingham and District Land Co v London and North Western Railway Co (1886) 34 ChD 261 (CA) 274–5: ‘a right to damages … is not a right to indemnity as such. It is theconverse of such a right. A right to indemnity as such is given by the original bargain betweenthe parties. The right to damages is given in consequence of the breach of the original contractbetween the parties. It is an incident which the law attaches to the breach of a contract, andis not a provision of the contract itself.’ Similarly, I think it is no coincidence that the basicmeasure of (compensatory) damages in contract is much the same as we find in the law oftorts (see, respectively, Robinson v Harman, above n 8, at 365 (Parke B) and Livingstone vRawyards Coal Co (1880) 5 App Cas 25 (HL) 39 (Lord Blackburn)). In both cases, the awardattempts to put the claimant in as good a position as she would have been in had there beenno breach, but, typically, in the latter case, there will be no agreement or undertaking betweenthe parties from which such an award might be said to derive.

31 Luxor (Eastbourne) Ltd v Cooper [1941] AC 108 (HL) 137 (Lord Wright); Shell UKLtd v Lostock Garage Ltd [1976] 1 WLR 1187 (CA) 1196 (Lord Denning MR); E Peel,Treitel: The Law of Contract, 12th edn (London, Sweet & Maxwell, 2007) 223, 229–36.

152 Charlie Webb



If this is true then it seems that any damages award can be regarded asprovided for by, and so as giving effect to, an implied term of the contract.This in turn appears to open up the possibility of supporting all suchawards by reference to the claimant’s right that the contract be performedto its terms. Now, this move is in itself unobjectionable, but its effect is tobring within the single notion of a right to performance a collection ofrights which have diverse bases and justifications. Some terms of thecontract, and the rights and duties to which they give rise, derive from theundertakings the parties have themselves given. Others terms, and so otherrights and duties, arise, despite the fact that the parties have not given anysuch undertaking, because some other principle of justice or efficiency orthe like is thought to justify it. I have been using the language of the rightto performance to cover only rights falling within the former class. Nodoubt we can extend the terminology of performance so that the latter setof terms, and correlative rights and duties, is also embraced.32 But thedifference between them remains. We just have to find different languageto express it.

My claim is that, whatever view we take of the basis of contractualobligation, if we recognise that contracts arise out of undertakings andlead to obligations to fulfil or make good those undertakings, the principlewhich supports such obligations (and their correlative rights) cannotaccount for whatever additional obligations the law recognises as owed byand to contracting parties that cannot be said to embody or to derive fromany such undertaking. Where the parties have made no undertakings as totheir rights and obligations upon breach, and performance of the undertak-ings they did make is no longer possible, any damages award cannot bejustified by reference to the undertakings the defendant did make and theprinciple which justifies the obligation to make good those undertakings.

This brings us to a second challenge to the position taken here, that theright to performance can support a damages award only where and to theextent that it provides the claimant with that performance, and hence thatwe must look elsewhere to justify damages awards where performance isno longer possible. The objection here is one recently expressed by ErnestWeinrib: that although the specific action required of the defendant maychange after breach—for example, to pay compensation rather than todeliver the shoes she undertook to provide—the duty, and its correlativeright, remain the same. As such, we should say that, although the

32 Though, so understood, the claimant’s right to performance becomes the right that thedefendant fulfil whatever obligations the defendant owes to her, or, pleonastically, the right tohave one’s rights effectuated. As such, the notion of a right to performance becomesessentially meaningless.




‘performance’ to which the claimant is entitled may change, it is theclaimant’s right to performance which is effectuated through, and so whichsupports, the damages award.33

Why should it be thought better to see these as two manifestations of thesame right, rather than simply different rights? For Weinrib this followsfrom a proper understanding of the nature of private law claims and, inparticular, of the relationship between rights and remedies. Weinrib arguesthat a claimant’s rights, and their infringement by the defendant, are notsimply conditions of the claimant’s claim and any remedy she is given, butare the reason for granting a particular remedy.34 The injustice of whichthe claimant complains when she brings a claim is that some right of hershas been infringed, and the point of the remedy—the order or award madeby the court—is to correct that injustice by realising or effectuating theclaimant’s right. On any other view, we must say that the injustice sufferedby the claimant, while a condition for granting some remedy, does notdictate the content of that remedy. Yet, if this is so, we have no reason formaking the availability of any given remedy dependent on the identifica-tion of a particular injustice.35

As I hope is clear already, I agree with much of this. I certainly believethat when faced with private law claims courts should be seeking toidentify the rights of the claimant and, where possible, to give effect tothose rights. Indeed, I would say that where there is no right of theclaimant to which the court can give effect, there is no basis for any awardor order in favour of the claimant beyond a declaration that her rights havebeen infringed. Nonetheless, I see no reason to accept Weinrib’s argumentthat, at least where ‘exact’ performance is no longer possible, breachchanges the content of the right to performance to a right to compensatorydamages, such that the right to performance lives on in and is effectuatedby such an award.

Firstly, Weinrib appears to assume that to reject his understanding ofsuch awards involves a disconnection of right and remedy and so leads tothe sort of incoherence described above. This would indeed be true if theaward could not be said to realise some right of the claimant. But this isnot what I am suggesting. My claim is simply that, where performance isno longer possible, a damages award cannot give effect to, and so besupported by, the claimant’s right to performance. A damages award maystill be justified if we can identify some other right of the claimant’s, onewhich is still capable of being effectuated by such an award. This is, Isuggest, the case with claims for compensatory damages. As such, there is

33 E Weinrib, ‘Two Conceptions of Remedies’ in C Rickett (ed), Justifying Private LawRemedies (Oxford, Hart Publishing, 2008) 3.

34 Ibid, at 15, 26.35 Ibid, at 7–8.

154 Charlie Webb



no reason for thinking that the logic or structure of private law demandsthat we view the right effectuated through compensatory awards as simplyanother manifestation of a claimant’s right to performance.

The question then is whether Weinrib’s is a better or more compellingaccount of a claimant’s rights than the one I am offering here. I haveargued that a given right can be said to be effectuated by, and so can itselfsupport, a particular award or order only where that order or awardprovides the claimant with the very factual advantages which that rightseeks to secure for her. Only then does the award give the claimant the verything or performance to which he or she is entitled. Weinrib, however,suggests that there are in fact two ways in which (primary) rights, such asthe right to performance, can be effectuated or ‘restored’.36 Firstly, andmore straightforwardly, there is what Weinrib terms ‘qualitative restora-tion’. These are the sorts of awards I have been describing, whereby theclaimant is provided with the exact thing or performance she is entitled to,as exemplified by orders of specific relief. The performance interestdamages awards I have argued for above would also fit in here. Secondly,Weinrib argues that a right may be effectuated in its ‘quantitative form’.Here, the claimant obtains not performance itself but the monetary valueof that performance. This is exemplified by standard awards of compensa-tory damages.

For Weinrib then, ‘quantitative restoration’, though giving the claimantsomething different from her initial right (or, at least, from her right asinitially formulated), is nonetheless to be regarded as a realisation of thatright. Why? It is of course true to say that we can put a value on rights andthe factual advantages they aim to secure for us. Accordingly, where a righthas been infringed and is not or cannot be specifically effectuated, we canstill ensure that the claimant receives something else of equal value to theperformance of which she has been deprived. Such a claimant will then notbe left worse off because of the breach. But this is not performance. As Iargued earlier,37 a claimant, upon entry into the contract, acquires a rightto performance irrespective of the value she places on that performance.The defendant’s duty is not to provide the claimant with something ofequal value to the performance for which the claimant contracted, nor is itsimply to ensure that the claimant is not left worse off by virtue of notreceiving that performance. It is a duty to perform, to provide theperformance the defendant undertook to provide. So, in so far as Weinribappears to be suggesting that there are two aspects or forms of theright—the qualitative aspect or form, focusing on the specific factualadvantages which the claimant can demand the defendant provide, and the

36 Ibid, at 13.37 See text accompanying nn 2–6.




quantitative, which looks to the value to the claimant of thoseadvantages—I think he is wrong.38 If a defendant breaches a contract orcommits a tort which happens not to leave the claimant worse off, we donot view this as only a partial breach, as a breach only of the ‘qualitative’aspect of the claimant’s right but not the ‘quantitative’. There are not twostrands to the claimant’s right. There is one indivisible performance towhich the claimant is entitled and which the defendant is under a duty toprovide.39

Weinrib’s answer would appear to be that, though this is all true whenthe right first arises, breach changes things.40 Breach changes, or at least

38 I also think that Weinrib’s account has a difficulty in accommodating damages forconsequential losses, that is, losses consequent upon and so additional to the loss inherent inthe defendant’s defective performance. Even if we do say that a claimant’s right toperformance has a quantitative form or can be viewed in quantitative terms, and hence that adamages award can be said to realise the claimant’s right to performance so understood, thiswould seem only to explain recovery of the value of that performance (or the differencebetween the value of that performance and the defective ‘performance’ in fact provided by thedefendant). The claimant’s right to performance is a right only to the specific factualadvantages which the defendant undertook to provide. Any further (consequential) advan-tages which the claimant may then derive from performance are not part of that performanceand hence are not part of, or protected by, the claimant’s right to performance. For example,say, I want a new crankshaft for my mill, and agree to buy one from you. Even though thatcrankshaft will, or may, enable me to make certain consequential gains, my right toperformance is simply a right that you provide me with that crankshaft. Accordingly, if myright to performance is to be assessed or understood quantitatively, it can cover only the valueof that crankshaft. If I am also to be able to recover the additional, consequential losses Isuffer as a result of not receiving the crankshaft, it cannot be on the basis that these are partof the performance to which I am entitled.

39 To similar effect, Arthur Ripstein (A Ripstein, ‘As if it Had Never Happened’ (2007) 48William and Mary Law Review 1957) has claimed that compensatory damage awardseffectuate a claimant’s primary right on the basis that such awards ensure that the claimantretains the ‘means’ (meaning the things or attributes—such as property and physical andmental powers—which are yours to employ in the pursuit of your ‘ends’) which the primaryright sought to secure for her. Awarding a sum of money to a claimant amounts to a provisionof such ‘means’. An appropriate sum will ensure that despite, for instance, being deprived ofthe shoes she was promised or despite the physical injuries the defendant caused her, theclaimant is left with no ‘fewer’ means—ie, that she is in as good a position to pursue her ends(ambitions, goals etc). But, as with Weinrib’s notion of quantitative restoration, this can beregarded as effectuating the claimant’s primary right only by stripping it of its specificcontent. So, rather than being understood as a right to a pair of shoes or to physical integrityetc, every primary right is treated as a right to the shoes/physical integrity or equivalentmeans. On this view, ‘means’ are essentially fungible. But when looking at contractualobligations, this does not reflect the content of the undertakings which we regard asgenerating those duties. Typically, my undertaking is to provide you with the shoes, not toprovide you with the shoes or any equivalent sum of money. Indeed, I think the same is trueof primary obligations throughout private law. They are not obligations to secure someabstract measure of means, but to provide us with certain specific factual advantages or toavoid certain specific factual harms. This also shows that Weinrib and Ripstein’s views are notfar removed from (and are open to many of the same criticisms as) the position taken byHolmes (see discussion above, nn 2–4).

40 Weinrib, above n 33, at 26.

156 Charlie Webb



can change, the content of the claimant’s right, and hence the ‘perform-ance’ to which she is entitled. But this appears impossible to square withany view of contract which sees the parties’ obligations as deriving fromand defined by reference to the undertakings they have made to oneanother. If (or to the extent that) we believe that contracts arise fromagreements or undertakings and involve obligations to fulfil those under-takings, the scope of those obligations will be determined, at least in thefirst instance, by the scope of those undertakings. So, where we contractfor you to sell me your shoes, my obligation to pay you derives from thatagreement, from my undertaking. The same goes for your obligation todeliver the shoes to me. But, unless you have explicitly or implicitlyundertaken this, the same cannot be said of any obligation you come underto pay me compensatory damages should you fail to give me the shoes. Toaccount for this obligation, and its correlative right, we must look beyondwhatever principle it is that supports holding us to the undertakings wehave made. So, though Weinrib argues that compensatory awards realise aclaimant’s right to performance, despite their different content, becausethey have a common ‘normative ground’,41 in fact it is by looking to therespective justificatory bases that we can best see why compensatorydamages awards should not be seen simply as manifestations of a claim-ant’s the right to performance.

Of course, if this is correct, it means that where performance is no longerpossible the claimant’s right to performance goes unrealised. One mightcontend that Weinrib’s argument, therefore, has the advantage of notforcing us to concede that, sometimes, rights will go unprotected and beleft uneffectuated. But, on this view, the right to performance is given effectonly by stripping it not only of its content but also what may be consideredits defining feature—its basis in the undertakings the defendant has madeto the claimant.

V. COMPENSATORY DAMAGES

This brings us at long last to awards of compensatory damages. As I haveargued over the preceding pages, these awards are best regarded asmanifesting and effectuating a right of the claimant’s which is distinct fromher right to performance. This right is commonly described as a ‘substitute’for the claimant’s right to performance,42 but I do not think this quitecaptures the nature of the right or its relationship to the right to

41 Ibid.42 See, eg, Photo Production Ltd v Securicor Transport Ltd [1980] AC 827 (HL) 848–9

(Lord Diplock); D Friedmann, ‘The Performance Interest in Contract Damages’ (1995) 111LQR 628, 629–31. For other uses of the language of substitute or substitutionary rights: seebelow nn 58–64 and accompanying text.




performance. The language of substitution suggests that, upon breach, theright to compensation for losses caused by the defendant’s failure toperform arises in place of the claimant’s right to performance. However,even where performance has been put beyond reach, I see no reason forconcluding that the claimant’s right to that performance has been lost.43

Certainly where performance remains possible we should continue toregard the claimant as entitled to that performance and our principalresponse to claims for breach of contract should be to ensure that thisperformance is provided. Moreover, claimants who have suffered a breachof contract (and courts faced with the task of responding to such claims)need not choose between demanding that the contract be performed andcalling on the defendant to make good the losses caused by non-performance. Hence, a claimant should be able to demand both that thecontract be performed, and that any losses which performance would notcorrect, for instance those caused by delay, be compensated.44 The onlylimit should be the usual rule that we do not allow double or inconsistentrecovery. As such, I think it is preferable to view the right to compensation,not as a substitute or replacement for the right to performance, but simplyas a distinct right with a distinct objective, and which exists and may beasserted alongside the right to performance.

What then is the content and basis of this right? Here I plan to say verylittle on why contracting parties should have, in addition to a right toperformance, a right to have losses caused by non-performance madegood. As I have argued already,45 not only is such a right distinct from theright to performance, but I think it must also be seen as resting on adistinct principle. My inclination is that, whatever this is, it is the sameprinciple which supports compensatory awards for torts and other wrongs.For present purposes, however, I shall leave this question to one side and,instead, take a closer look at the loss which compensatory awards seek tomake good.

The language of ‘loss’ tends to be used in two different, though related,senses. Firstly, we sometimes use the word ‘loss’ to describe certain specificphysical harms or unwanted occurrences. In this way, a broken leg,smashed window or an undelivered pair of shoes can all be described aslosses. Understood in this way, loss is inherent to breaches of contract,indeed to all breaches of duty. Every time there is a breach of contract, theclaimant ends up being deprived, if only temporarily, of some part of the

43 See text accompanying nn 60–61.44 This indeed is the position the law currently takes whereby a claim for specific

performance or an action for an agreed sum can be combined with a claim for compensatorydamages: see, eg, s 49 of the Supreme Court Act 1981; A Burrows, Remedies for Torts andBreach of Contract, 3rd edn (Oxford, Oxford University Press, 2004) 506.

45 See text accompanying n 32.

158 Charlie Webb



performance, of something—an asset, service and so on—to which she isentitled. In simple terms, breach means that, as a matter of factual reality,the claimant does not get something she should have got.46

Now, if were are to understand ‘loss’ as specific physical harms oroccurrences, losses can be made good, if at all, only by ensuring that therelevant harmful or unwanted state of affairs no longer obtains. So if theloss of which the claimant is complaining is a broken window or notreceiving a pair of shoes, then that loss can be corrected only by seeing thatthe window is repaired and so left ‘unbroken’ or by ensuring that the shoesare delivered. When loss is understood in this way, a claim to have that lossmade good is simply another way of describing a claim that the contractshould be performed and hence that the claimant should be provided withthe very factual advantages which the defendant undertook to provide andto which she is entitled. As such, these ‘losses’ are reversed by orders ofspecific relief and, as suggested earlier, awards of damages which enablethe claimant to obtain performance from some third party. It also followsthat where performance is no longer possible, and hence where the specificfactual advantages which performance would provide can no longer beobtained, then the loss of those advantages can no longer be corrected andhence cannot form the basis of a claim.

There is, however, a second use of the word ‘loss’. This describes not thefactual, physical harms or adverse occurrences which we may suffer but,broadly, the impact they have on us, on our happiness or quality of life. Inthis sense we speak of the broken leg, smashed window or undeliveredshoes as a source of loss or of the loss flowing from such occurrences. Sinceloss here is not identified as specific physical harms we may suffer but isinstead something consequent upon them, its reversal is not dependentupon and so need not involve the reversal of those specific physical harms,and indeed such a reversal is possible even where those harms can nolonger be made good. Rather, if loss is conceived as a worsening of theclaimant’s position, a reduction in her quality of life, then making goodthat loss requires that we do something for the claimant which improvesher quality of life sufficiently to offset the reduction brought about by thedefendant’s breach. It is loss in this sense which I believe is the proper focusof compensatory awards and which provides the basis of an award ofdamages even where performance of the defendant’s primary obligations isno longer possible.

Though losses will always result from real world occurrences, the lossitself is distinct from those events. We suffer a loss every time something

46 This is not quite the same as saying that the breach of contract is itself the loss, since, onthis understanding, loss describes the factual impact on the claimant of the defendant’sconduct (eg, not receiving the promised pair of shoes, an insufficiently deep swimming pool),whereas breach describes the legal ‘quality’ of that conduct.




happens—we experience or receive something—which we value less thanthe state of affairs which existed before or which would otherwise havecome about (depending on our point of comparison). So, when a claimantwho has suffered a breach of contract complains of having suffered a lossin this sense, her complaint is not (simply) that she has not received thefactual advantages which performance would have brought but that,because she has been deprived of those advantages, she is now in a worseposition—one she values less—than that which she would have occupiedhad performance been forthcoming. Here then loss is not an inevitablefeature or consequence of a breach of contract. A claimant will suffer a lossonly where and to the extent that she values performance and the factualadvantages it would bring her more than the defective ‘performance’ whichthe defendant in fact provided.

This has a number of important consequences. Firstly, the measure, andindeed the very existence, of a loss follow from and will be determined byour preferences and objectives. Because these vary from person to person,loss is necessarily personal or subjective. What I regard as a loss is likely tobe different from what you consider to be a loss since we have differenttastes and goals and hence value things differently. For this reason thecourts are correct to draw attention to the ‘consumer surplus’; thepossibility that, and the extent to which, a claimant places a value onperformance in excess of its market value.47

Secondly, loss is not co-extensive with, and so does not depend on, areduction in the funds or assets at our disposal. We clearly can and dovalue assets, and a loss of assets is one possible source of loss. However, itis clear that there are other ‘things’ which can also make a contribution toour ability to satisfy our wants and fulfil our objectives, and hence towhich we also attach value, such as experiences, knowledge and relation-ships. It is because of this that we do not find it counter-intuitive thatpeople choose to pay money for holidays or to see a film or even makedonations to charity. By making such choices, we are saying that we valuethe experience of the holiday or film, or the welfare of the beneficiaries ofthe charity, more highly than the money itself (and the other ‘things’ whichthat money could provide for us). By contrast, if we were to consider loss

47 See, eg, the speech of Lord Mustill in Ruxley Electronics and Construction Ltd vForsyth, above n 8 and the speeches of Lords Scott and Hutton in Farley v Skinner [2001] 3WLR 899 (HL); see, too, D Harris, A Ogus and J Phillips, ‘Contract Remedies and theConsumer Surplus’ (1979) 95 LQR 581. One might even suggest that the language of theconsumer surplus is misleading both in that it suggests that, when measuring loss, our startingpoint should be market value (with the consumer surplus being a deviation from andexception to this measure), and that the relevance of subjective value is limited to consumercontracts. As a matter of principle, loss is always subjective and market value is only ever anindication of what the claimant’s (subjective) loss may be. In practice, however, given thedifficulty of establishing the value the claimant places on performance, it is understandablethat courts should use market value as their benchmark.

160 Charlie Webb



as dependent upon and co-extensive with a reduction in our assets orwealth, we would be forced to conclude that all contracts for the provisionof a service, or any other contract which does not leave the claimant withsome ‘marketable residuum’,48 necessarily leave the claimant worse off. Abreach of such a contract could cause a loss to the claimant only if (andthen only to the extent that) she paid in advance or parted with otherassets in reliance on the defendant’s performance. Similarly, such a restric-tive understanding of ‘loss’ leaves no room for recovery in respect ofpersonal injuries or illness, save in so far as these lead us to incur out ofpocket expenses.

More fundamentally, we can challenge the preconception that lossesstemming from a reduction in one’s assets are of a different order or kindfrom those flowing from, for example, a lousy holiday, disrupted night’ssleep or twisted ankle.49 If we ask why it is that we attach value to assets,the answer is that, through their use, we are (or may be) better able topursue our goals and satisfy our wants. This may seem to contradict theview that an asset’s value is determined by what others are willing to payfor it. However, not only is sale among the ‘uses’ to which one can putone’s assets, but, more importantly, the reason we are all willing to pay forassets is precisely because we value the uses to which those assets can thenbe put. As such, market value follows from and is a reflection of the valuepeople generally attach to use and enjoyment of the asset. An asset which isno use to anyone will have no market value.

Accordingly, we value assets in the same way and for the same reasonthat we value experiences, knowledge, relationships and the like; becauseof the contributions they make to our quality of life, to our ability toachieve our goals and satisfy our wants. Consequently a reduction of ourassets is a loss precisely because it inhibits the pursuit of those goals and soreduces our quality of life. So, while we can certainly distinguish losses inthe sense of specific factual harms—property damages, physical harm,psychological injury and so on—once we understand loss as somethingconsequent upon, and hence distinct from, those harms, we must under-stand such losses as necessarily fungible.

48 This expression is used in unjust enrichment literature to distinguish services whichleave the recipient with some increase in the assets or wealth at her disposal, and ‘pure’services which do not: see J Beatson, ‘Benefit, Reliance, and the Structure of UnjustEnrichment’ in The Use and Abuse of Unjust Enrichment (Oxford, Clarendon Press, 1991)29–32. The understanding of ‘loss’ put forward here is in substance the mirror image of thenotion of ‘benefit’ or ‘enrichment’ employed by those who argue that pure services can beenriching.

49 As manifested in the distinction commonly drawn between ‘financial,’ ‘economic’ or‘pecuniary’ losses and ‘non-pecuniary’ losses: see, eg, Burrows, above n 44, at 29; P Cane, TheAnatomy of Tort Law (Oxford, Hart Publishing, 1997) 98–9; Wright v British RailwaysBoard [1983] 2 AC 773 (HL) 777 (Lord Diplock); Heil v Rankin [2001] QB 272 (CA) 293(Lord Woolf MR).




This may all sound very different from the way the notion of ‘loss’ istypically presented in the cases and texts. Nonetheless, I do not think thereexist any great differences of substance between those accounts and myown. In particular, the fact that we award compensatory damages in caseswhere performance is no longer possible, and hence where the relevantfactual physical harm suffered by the claimant cannot be put right, showsthat we do not equate loss with such factual harms. It is clear, however,that the courts have, on occasions, lost sight of the fact that losses canoccur other than through a diminution of or damage to the assets at theclaimant’s disposal.

Take, for instance, situations in which A contracts with B for some assetor service to be provided to C. Here, the problem has been seen to be that,if B breaches, it is C and not A who suffers the loss, since it was C and notA to whom the asset or service was to be supplied. This then appears tolead to the result that A is unable to recover anything more than nominaldamages, which, when combined with the common law rule that C, whohad suffered a loss, had no claim, meant that the law let breaches of suchcontracts go largely unremedied.

In fact, we have two clear bases for awarding A substantial damages insuch cases.50 Firstly, A has a right to performance of that contract, and so,as I argued earlier, we can justify awarding her damages assessed on a costof cure basis so as to enable her to purchase an equivalent ‘performance’from another source. Secondly, and alternatively, A can say that she hassuffered a loss as a result of B’s breach of contract. The fact that A wasprepared to pay B to provide the relevant service or asset to C, shows thatA placed a value on that performance. Where that performance is notforthcoming, A is accordingly deprived of something she values. Thedefendant’s breach has led to a state of affairs—C not receiving therelevant asset or service—which A values less than the state of affairswhich should have eventuated—proper performance and C’s receipt of theintended asset or service. This is a loss. We are led to the conclusion thatthere is no loss here only by adopting a conception of loss which looksonly to the quantum and value of the assets at the claimant’s disposal. Aswe have seen, such a restricted notion of loss is insupportable.

Another example of the courts making heavy weather of ‘loss’ isprovided by the line of cases exemplified by Wrotham Park v Parkside

50 A failure to distinguish between these two alternative bases for an award of damages isevident in the various judicial attempts to devise a principled solution to the problem posedby these cases. See, eg, the dicta of Lord Scarman in Woodar Investment Development Ltd vWimpey Construction UK Ltd [1980] 1 WLR 277 (HL) 300–1 and Lord Griffiths in LindenGardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85 (HL) 97, and theapproaches favoured by Lords Goff, Browne-Wilkinson and Millett in McAlpine v Panatown,above n 30. See generally Webb, above n 3, at 53–6.

162 Charlie Webb



Homes.51 Here the defendant built on its land in breach of a restrictivecovenant. The claimant, to whom the obligation was owed, sought aninjunction to restrain further building and for the demolition of thebuildings already erected. Brightman J, however, refused to order theirdemolition on the basis that this would be a waste of much neededhousing. The claimant was, however, entitled to damages and, though thedefendant’s breach had not affected the value of the claimant’s land,Brightman J awarded the claimant damages of £2,500, on the basis thatthis was the sum of money which ‘might reasonably have been demandedby the [claimant] from [the defendant] as a quid pro quo for releasing thecovenant’.52

It seems tolerably clear that Brightman J considered this sum compensa-tion for a loss caused by the defendant’s breach of duty. However, thisunderstanding of the award has come in for considerable criticism overrecent years. The supposed flaw in the compensatory analysis of cases suchas Wrotham Park is that the claimant in fact suffers no loss, and, as such,there is nothing to compensate. Focusing on Brightman J’s reference to thefee that the claimant could have demanded to release the defendant fromits obligation,53 the standard objection is that this can only be viewed as agenuine loss where the claimant would in fact have been willing and ableto obtain such a sum in return for granting a release to the defendant.Where, as in Wrotham Park,54 no such release would have been granted,any such ‘loss’ is routinely dismissed as a fiction.55

No doubt it is a fiction to say that the claimant has lost a sum of moneywhich it would not in any case have obtained. Nonetheless, we should haveno difficulty identifying a loss on such facts. The claimant did not want thedefendant to build on its land. When the defendant, in breach of theobligation it owed to the claimant, did build, it brought about a state ofaffairs which the claimant wished to avoid, a state of affairs which itconsidered worse than the state of affairs which obtained previously andwould have persisted had the defendant abided by the restrictive covenant.This is a plain loss. The fact that the claimant would not have released thedefendant from its obligation, even in return for a payment of thousands ofpounds, far from amounting to a denial of loss, patently supports such afinding. The only question then is how we measure that loss. Given thatthe courts have traditionally been reluctant to stray from valuation

51 Wrotham Park Estate Co Ltd v Parkside Homes Ltd [1974] 1 WLR 798 (Ch D).52 Ibid, at 815.53 See, too, R Sharpe and S Waddams, ‘Damages for Lost Opportunity to Bargain’ (1982)

2 OJLS 290; Jaggard v Sawyer [1995] 1 WLR 269 (CA); cf Attorney-General v Blake, aboven 30, at 281 (Lord Nicholls).

54 It was found that the claimant would not have consented to any such release.55 See, eg, A Burrows, The Law of Restitution, 2nd edn (London, Butterworths, 2002)

477; Edelman, above n 1, at 99–101.




determined objectively by reference to the market, it is entirely unsurpris-ing that they have approached this question by asking what sum theclaimant could reasonably have demanded in return for releasing thedefendant from her obligation, since the price which a hypotheticalclaimant would accept in return for allowing the defendant to ‘breach’identifies the value the claimant places on the defendant abiding by herobligation. To object that this is not the sum of money which the claimantwould have charged is simply to object to adopting an objective, market-based assessment of loss, rather than one which takes account of theclaimant’s own priorities and preferences. As I have argued earlier,56 inprinciple, loss should indeed be viewed, and hence identified and quanti-fied, on a subjective basis. But this just means that our assessment of loss inthese cases has been defective, not that a compensatory analysis must beregarded as fictitious and so to be rejected.

Once we understand that a loss is properly to be regarded as a reductionin the claimant’s quality of life, we can see that the point of compensatoryawards is to do something for the claimant which she values and so whichimproves her quality of life, such that, as a result of the award she ends upin as good a position as she would have occupied had the contract beenperformed. In this way, though the claimant has been deprived of one thingshe values—performance of the contract—she gets something else whichimproves her quality of life by an equivalent degree. As I have beenstressing over the previous paragraphs, loss should not be viewed asdependent on and co-extensive with a diminution in the assets at theclaimant’s disposal. As such, when it comes to making good losses, there isno reason for such an award to involve the payment of money or thetransfer of some other asset to the claimant. In principle, providing theclaimant with any ‘thing’ she values equally to performance (and anyconsequential losses) would suffice. Nonetheless, money clearly providesthe most convenient means of compensation. Though individual’s tastesand priorities differ, money is necessarily valuable to all of us, since itprovides a ready means through which we can acquire those specific‘things’ we want or need to pursue our objectives.

This then explains why we are right to leave claimants free to use themoney they receive as compensatory damages as they wish. Unlike awardsjustified on the basis of the claimant’s right to performance, the aim ofcompensatory awards is not to correct the specific harms caused by thedefendant’s breach but rather to ensure that the defendant does not leavethe claimant worse off as a result of that breach, and hence as a result ofsuffering those specific harms.57 A money payment does this because it

56 See text accompanying n 47.57 It is because of this that suggestions that compensatory awards for ‘non-pecuniary’

losses are necessarily artificial or can only be based on convention are misguided: see, eg,

164 Charlie Webb



allows the claimant to purchase other assets or experiences which shevalues, and so which bring about an improvement to the claimant’sposition sufficient to offset the worsening of her position caused by thedefendant. Accordingly, there is no sense in placing any limits on theclaimant’s use of the money she receives as compensation. Indeed, if theaward is to have it intended effect, the claimant must be free to spend themoney as she likes.

VI. SUBSTITUTIVE DAMAGES

Finally I want to say a little about an argument which is gaining a fairamount of support among commentators and which, if true, wouldprovide a further justification for an award of substantial damages in theevent of a breach of contract. This is that damages are and should beavailable as a substitute for the right infringed by the claimant, with suchdamages then being labelled ‘substitutive’ damages.58

The argument is that where a claimant’s right has been infringed sheshould be entitled to damages from the defendant assessed by reference tothe ‘value’ of that right, measured objectively and at the date of thebreach.59 Such awards do not depend on showing that any loss has been

Burrows, above n 44, at 29; Wright v British Railways Board, above n 49, at 777 (LordDiplock). If we are not seeking to put right the specific physical harms suffered by theclaimant, then all losses are fungible, since they all lie in the impact of the defendant’s conducton the claimant’s overall quality of life. Though evidential difficulties may make it harder toestimate the gravity of this impact in some cases than in others, in principle all losses,whatever their source, can be assessed in monetary terms. This is not to say that claimantsshould always be entitled to recover the full sum of money it would take to bring about animprovement to their quality of life sufficient to offset the reduction caused by the defendant.Other considerations and principles may well justify restricting recovery to a lesser sum. Inparticular, where a claimant has been caused serious injury or illness, the impact on herquality of life will usually be so great that it can be offset only by the payment of a huge sumof money. Requiring a defendant to hand over such a sum would often be financially cripplingand, at least where any blame to be attached to the defendant is minimal, may be consideredunjust: see, eg, Phillips v The South Western Railway Company (1879) 4 QBD 406, 407(Cockburn CJ). However, we should view these and similar considerations as going not to theidentification or measurement of the claimant’s loss but to the necessarily secondary questionof how much of that loss the defendant should be required to bear.

58 The language of ‘substitution’ and of ‘substitutive’ (or ‘substitutionary’) damages hasalso been used to describe other types of award. Some have used it to describe conventionalcompensatory awards (see the references above n 42). Smith, above n 10, identifies as‘substitutionary damages’ awards which I have referred to here as ‘performance interest’damages. It is important to distinguish these uses of the language of ‘substitution’ (and theawards they describe) from that discussed in the text.

59 See, eg, R Stevens, Torts and Rights (Oxford, Oxford University Press, 2007) 59–84; MMcInnes, ‘Gain, Loss and the User Principle’ (2006) 14 Restitution Law Review 76; cf SElliott and C Mitchell, ‘Remedies for Dishonest Assistance’ (2004) 67 MLR 16; J Edelman,‘Gain-Based Damages and Compensation’ in A Burrows and Lord Rodger (eds), Mapping theLaw: Essays in Honour of Peter Birks (Oxford, Oxford University Press, 2006) 141, 153–8;




caused to the claimant, and indeed exist independently of any claim theclaimant may also have for the recovery of such losses. Instead they followsimply from the infringement of the claimant’s rights.

A number of points can be made. Firstly, we can ask why we should beseeking to a put a value on the claimant’s rights. The language of ‘value’ isusually employed when we are looking to identify and assess losses andgains, as when we are concerned with awarding compensatory damages forlosses caused by the defendant’s breach. Yet such awards are heralded asnoteworthy and distinctive precisely because they are different from, andexist independently of, standard compensatory claims. It is claimed thatsubstitutive damages are awarded not to make good the consequences ofthe defendant’s infringement of the claimant’s right but simply as areflection of the right that has been violated. Where then does value comeinto it?

One answer would be that substitutive awards are aimed at compensat-ing a loss; it is just a different sort (or sense) of loss from that which is thefocus of ‘standard’ compensatory awards.60 So, rather than seeking tocompensate the factual losses flowing from the defendant’s breach, substi-tutive awards seek to make good the loss of the claimant’s legal rightsinherent in that breach. On this basis, we are looking to value theclaimant’s right precisely because this is the loss the claimant has suffered.

However, I doubt whether such a reformulation of the claimant’s loss issuccessful. Firstly, we can question the premise that an infringement of aclaimant’s right entails or results in a loss of that right. Though breach cancause us to lose the factual advantages which rights seek to secure for us, Ido not think it is quite right to say that we thereby lose the rightsthemselves. I think a better account is provided by Lord Clyde in McAlpinev Panatown:

[W]hile frustration may destroy the rights altogether so that the contract is nolonger enforceable, a failure in the obligation to perform does not destroy theasset. On the contrary it remains as the necessary legal basis for a remedy. Afailure in performance of a contractual obligation does not entail a loss of thebargained-for contractual rights.61

So, where a duty is breached, and hence a right infringed, far fromregarding that right as lost, the law’s principal response should be to give

D Pearce and R Halson, ‘Damages for Breach of Contract: Compensation, Restitution andVindication’ (2008) 28 OJLS 73. For further analysis and criticism of this argument: see ABurrows, ‘Are “Damages on the Wrotham Park Basis” Compensatory, Restitutionary, orNeither?’ in D Saidov and R Cunnington (eds), Contract Damages: Domestic and Interna-tional Perspectives (Oxford, Hart Publishing, 2008) 165.

60 See McInnes, ibid, at 84–6; cf Stevens, ibid, at 61–2, 78.61 McAlpine v Panatown, above n 30, at 534. See too Weinrib, above n 33, at 12–13 and

Ripstein, above n 39, at 1971–72, 1978–82.

166 Charlie Webb



effect to that right, whether through specific enforcement or through adamages award such as that marked out above. And even where perform-ance is no longer possible, I do not think that it follows that we shouldview the right as having been extinguished, as opposed, for instance, toconcluding that it simply has no further bearing on the parties’ futureconduct.

In any case, even if we do consider that it is correct to say that theclaimant’s rights have been lost or damaged, there is then the question ofhow we value that loss. The law recognises us as having particular legalrights precisely because it considers that we should obtain certain factualadvantages (and avoid certain factual harms or disadvantages). So we havea right to performance of contracts because the law wants to secure for usthe factual advantages of such performance. But then the obvious measureof the claimant’s lost right is by reference to the factual advantages ofwhich she has been deprived by reason of the defendant’s failure to respectthat right. However, if this is true then the argument for substitutivedamages does not justify awarding the claimant any more than the sum towhich she is entitled straightforwardly as compensation for her factuallosses. Conversely, if, as claimed, substitutive damages are to justify anaward in excess of the claimant’s factual losses, we need to be able to saywhy we should attribute a value to the claimant’s right to performance inexcess of the value of the factual advantages which that performancewould secure. Simply to say that substitutive damages are assessed objec-tively and at the date of breach, whereas compensatory damages in respectof factual losses are to be assessed subjectively and at the date of judgment,and that this is why they can give us different figures, only prompts thequestion of why we should adopt different modes of assessment in eachcase.

A second version of the substitutive damages argument does not seek toexplain such awards on the basis that they reflect any sort of loss to theclaimant. But then we have no obvious answer to the question of why weshould be concerned with putting a value on the claimant’s right toperformance. A court is confronted by a claimant who has suffered abreach of contract; her rights have clearly been infringed. But how do weget to the conclusion that the court should then award the claimant thevalue of that right? To say that damages should be ordered as a ‘substitute’for this right or to put the claimant in the ‘next best’ position to actuallyreceiving performance gets us no nearer to an answer.62 In what way issuch an award of damages a substitute for performance? Why should weconsider this, rather than some other response, the next best way of dealingwith breach?

62 Cf Stevens, above n 59, at 59–60.




The fact that the claimant has a right to performance gives us ajustification for an award of damages where and to the extent that thisgives her the performance to which she is entitled. The claimant can claimthat sum because that gives genuine effect to her right. She gets the veryfactual advantage—performance—that the right exists to secure for her.That cannot be said of a substitutive damages award of the kind describedhere. Nor can such an award be justified on the basis that it ensures thatthe claimant is not left worse off as a result of not receiving theperformance to which she is entitled, since proponents of substitutivedamages avowedly claim that such awards permit recovery in excess of theclaimant’s real, factual losses.

Since substitutive damages neither give claimants the performance whichthey are due and for which they have contracted, nor make good lossescaused by the defendant’s breach, we are left in search of any justificationfor them. The basic question is simple: why should claimants be entitled torecover a sum of money which is in excess of their losses and which in noway gives them the performance to which they are entitled? I am notintending to suggest that we cannot justify damages awards other than onthe two bases I have put forward here.63 My claim is simply that any suchaward must be supported on some basis other than that it vindicates oreffectuates the claimant’s right to performance.

One possible argument is that, beyond the question of protecting theclaimant’s rights, breach of contract threatens the very practice of contract-ing and the various broader moral or political values which contract lawmay be thought to support and embody. This may be what is meant bythose who argue that we need to take performance, and the right toperformance, of contracts seriously, or that an award of damages shouldreflect the seriousness of the right.64 However, if we want an award ororder that amounts to a clear statement that the claimant’s rights have beeninfringed then this can be done by a declaration or an award of nominaldamages. If we want to send out a stronger message that contracts are animportant social and legal institution, and hence that contractual rightsshould be respected, then this may give us a further ground for an award ofsubstantial damages. If so, the assessment of such an award should dependon a consideration of how this message may best be conveyed. There is noreason to think that this sum should bear any relation to the value which

63 In particular, I have said nothing on the question of if and when we can justify damagesawards assessed by reference to the gains made by the defendant as a result of her breach. Theanalysis presented here bears on this issue only to the extent that it tells us that such awardscannot be justified on the basis that they give effect to the claimant’s right to performance.Whether some other justification for such awards can be found is a question I leave open.

64 Cf Stevens, above n 59, at 78–9, 84, argues that the proper measure of substitutivedamages turns upon the ‘seriousness’ of the infringement of the claimant’s right.

168 Charlie Webb



may be attached either to the performance the claimant is due or to theright which aims to secure her that performance.

VII. CONCLUDING REMARKS

I think there is an important lesson here for those of us who are concernedwith the justification of the awards and orders granted by courts not onlyin relation to claims following breaches of contract but in any aspect ofprivate law. The recognition of legal rights follows from some understand-ing that the claimant deserves to be accorded certain factual advantages (orto be spared certain factual harms). The claims the law supports and theorders and awards courts may make then seek to ensure, and are justifiedto the extent that they do ensure, that the claimant then obtains the veryfactual advantages which her rights are designed to secure for her. If we areto argue that the law should recognise a certain type of claim oraccommodate a certain type of award, we need to be able to say what rightof the claimant such a claim or award effectuates. This requires us to showhow such a claim or award is effective in bringing about the factualadvantages which that right seeks to secure. If we are, in turn, to be able todefend a particular view of what rights the law should accord to individu-als we need to be able to say why the law should be concerned withsecuring the relevant factual advantages for the identified class of rightholders. Any such argument necessarily takes us into the fields of morality,economics, politics and the like—in essence, wherever we consider that thenorms and ideals which shape and justify the law are to be found.

In this sense, this article has only touched the surface of the question ofhow the law should respond when faced with claims following breaches ofcontract. Much work needs to be done in explaining why a claimantshould be accorded a right to performance, and indeed in examiningwhether this is a necessary consequence of all contracts. Nor have I saidanything on the question of the principles which may challenge therecognition and protection of a right to performance, and hence whichmay limit the claims we make available to contract claimants.65 For now, itis enough to say that recognition of a right to performance requires that wedo not limit claimants to recovery of losses caused by the defendant’sbreach, but actually give them, where possible, the performance to whichthey are entitled. This can be done through specific enforcement of thecontract. Alternatively, the right to performance can, sometimes, beeffectuated through an award of damages which the claimant uses to

65 Here we may also note that there are clearly principles, such as those underlying therules on remoteness and mitigation, which limit the scope of the claimant’s right to recovercompensation for losses caused by the defendant’s breach.




purchase an equivalent ‘performance’ from an alternative source. If we areto argue for awards beyond these, we must look elsewhere for ourjustification.

170 Charlie Webb



7

Damages and the Right toPerformance: A Golden Victory or

Not?

ROBERT STEVENS

I. INTRODUCTION

A STORY. When I was to be married, the task of purchasing thewedding rings fell to me. I purchased a pair of rings from a jewellerin London’s Hatton Garden, a gold ring for me and a ring sold to

me as made of platinum for my wife to be.Over the years my wife’s wedding ring changed colour. Instead of the

steely-white of platinum it became yellowy gold. She took it to anotherjeweller who explained that her ring was not platinum but gold, plated inrhodium so that it looked like platinum. It was worth much less than thering I had thought I was buying.

Far from being upset, my wife was pleased. She thought that the cheaper‘defective’ wedding ring much more accurately represented the state of ourrelationship than any upmarket ring. She also, I suspect, likes it both as atalking point and as a reminder to me that I am not as smart as I like tothink I am. Her happiness makes me happier.

In law, what remedy is available to me against the jeweller (who I shallassume made an honest mistake)? I cannot get my money back: my wifewould never be parted from her wedding ring and would not now want thesort of ring I had intended. I am not in any way factually worse off as a resultof the jeweller’s breach of contract. The capital or re-sale value of the ring islower than the one I was promised, but the ring is my wife’s not mine, and wasinitially given by the jeweller to my best man to hold for her, not to me. In factthe ring as sold has left both me and my wife happier and factually better offthan if I had had the ring I was promised. It has even proved unexpectedlyuseful in my job of teaching and explaining the law.

Even though I have suffered no actual loss I am still entitled to damages,assessed as the difference in value at the time of performance between the



ring I was promised and the ring delivered.1 This article seeks to defendthis result on the basis that damages are awarded, in the first instance, tovindicate the right violated, here the contractual right to a wedding ring ofa particular quality. Such damages need to be differentiated from damagesawarded to compensate for loss consequent upon the breach of a contract.The failure to differentiate between damages awarded as a substitute forthe contractual right and those awarded to compensate for consequentiallosses suffered, leads to confusion and muddle.

II. WRONGS

A breach of contract is a wrong. A wrong is a breach of a duty. A breach ofa duty is the infringement of a right. That contracts give rise to primaryrights of performance can be demonstrated in a number of ways. Forexample, if a contract only gave rise to a right to performance or a right todamages for failure to perform, there would be no doctrine of frustrationbecause it is never, or almost never, impossible to pay a sum of money byway of damages. The remedies which compel the performance of theprimary contractual obligations (injunctions, specific performance, and byfar and away the most common remedy of all, the action for the agreedsum) similarly disclose the existence of the primary right to performance.The one and only ‘interest’ which a contract gives rise to is the ‘interest’ orright to counter-performance.2 The infringement of the primary right toperformance gives rise to a secondary right to damages which did not existprior to breach.3 The existence of damages whose role is to vindicate theright to performance which has been infringed, and not to compensate forloss suffered as a result of the infringement, is one way of revealing theimportance attached to the performance of what has been promised.4

Many other wrongs entitle the victim to claim substantial damages byway of substitution for the right infringed.5 ‘[E]very violation of a rightimports damage.’6 Two examples will be given here.

1 Sale of Goods Act 1979, s 53(3).2 D Friedmann, ‘The Performance Interest in Contract Damages’ (1995) 111 LQR 628;

contra C Webb, ‘Performance and Compensation: An Analysis of Contract Damages andContractual Obligation’ (2006) 26 OJLS 41.

3 Photo Production Ltd v Securicor Transport Ltd [1980] AC 827 (HL).4 See also D Pearce and R Halson, ‘Damages for Breach of Contract: Compensation,

Restitution and Vindication’ (2008) 28 OJLS 73.5 R Stevens, Torts and Rights (Oxford, Oxford University Press, 2007) 59–91. See also

Lord Scott, ‘Damages’ [2007] Lloyd’s Maritime and Commercial Law Quarterly 465.6 Neville v London Express Newspaper Ltd [1919] AC 368 (HL) 392 (Viscount

Haldane). See also Nicholls v Ely Beet Sugar Factory Ltd [1936] 1 Ch 343 (CA) 350 (LordWright MR); Jones Brothers (Hunstanton) Ltd v Stevens [1955] 1 QB 275 (CA) 280 (LordGoddard CJ).

172 Robert Stevens



The first, from 1703, is Ashby v White.7 In that case, the claimant wasprevented from voting by a constable on the pretext that he was not asettled inhabitant. The claimant’s preferred candidate was elected and so hehad no loss of any kind consequential upon the infringement of his right tovote, except perhaps some bruised feelings. In the Court of King’s Bench amajority rejected the claim. Holt CJ dissented:

[S]urely every injury imports a damage, though it does not cost the party onefarthing, and it is impossible to prove the contrary; for a damage is not merelypecuniary, but an injury imports a damage, when a man is thereby hindered ofhis right.8

On appeal to the House of Lords, his dissent was upheld by a majority oflay peers. Harvey McGregor describes the approach of Holt CJ as‘fictional’.9 However, without it, it is impossible to understand why thecommon law allows damages to be recovered without the proof of loss tothe claimant or gain to the defendant.

The law has not settled on one label for this substitutive award.Sometimes the distinction between substitutive damages and consequentialloss is described as that between ‘general’ and ‘special’ damages.10 How-ever, these terms are used in a variety of ways and are best avoided.11

Again, sometimes the substitutive award is termed ‘direct loss’ and iscontrasted with ‘consequential loss’.12 The label ‘direct loss’ is misleading,however, as the substitutive award is available even though no loss issuffered as a matter of fact. Substitutive damages are not compensatory forloss, properly so-called, at all.

Perhaps the most famous example in the law of torts is the decision ofthe House of Lords in The Mediana.13 The plaintiffs operated a port. Theirlightship was damaged as a result of the defendant’s carelessness. Theplaintiffs had a spare lightship to take the place of any ship damaged and

7 Ashby v White (1703) 1 Bro PC 62, 1 ER 417 (HL), reversing (1703) 2 Ld Raym 938,92 ER 126 (KB).

8 Ibid, at 137.9 H McGregor, McGregor on Damages, 17th edn (London, Sweet & Maxwell, 2003)

360.10 Ratcliffe v Evans [1892] 2 QB 524 (CA) 528 (Bowen LJ).11 Cf J Jolowicz, ‘The Changing Use of “Special Damage” and its Effect on the Law’

[1960] CLJ 214; McGregor, above n 9, at 19–23.12 Burdis v Livsey [2003] QB 36 (CA) [87].13 Mediana v Comet (The Mediana) [1900] AC 113 (HL). See also Steam Sand Pump

Dredger v Greta Holme (The Greta Holme) [1897] AC 596 (HL); Mersey Docks andHarbour Board v Owners of SS Marpessa (The Marpessa) [1907] AC 241 (HL); AdmiraltyCommissioners v SS Chekiang (The Chekiang) [1926] AC 637 (HL); The West Wales (1932)43 Ll L Rep 504; Edmund Hancock (1929) Ltd v ‘Ernesto’ [1952] 1 Lloyd’s Rep 467 (CA);Lord Citrine v Hebridean Coast (The Hebridean Coast) [1961] AC 545 (HL) [The HebrideanCoast]; Birmingham Corporation v Sowsbery [1970] RTR 84; cf Alexander v Rolls-RoyceMotor Cars Ltd [1996] RTR 95 (CA); Bella Casa Ltd v Vinestone Ltd [2005] EWHC 2807(TCC).

Damages and the Right to Performance 173



so were not forced to obtain a replacement whilst the damaged ship wasrepaired. There was no dispute that the defendants were liable to compen-sate the claimants for the cost of repairs, but were they additionally liablefor the loss of the use of the lightship during the period of repairs?Although the loss of the use of the ship that had resulted had not left theclaimant factually worse off, damages were payable representing the valueof the right to use the ship during the period of repair. It is sometimessuggested that the principle in The Mediana should be confined tonon-profit-making assets,14 but the distinction between profit-making andnon-profit-making assets was rejected by Lords Reid and Morton in TheHebridean Coast15 and is inconsistent with the law as we find it.16

Whether the right is wholly commercial is relevant, as we shall see, to itsquantification, but this does not mean that commercial parties are confinedto claims for consequential loss.

Why are damages awarded without proof of loss? The answer is thatdamages awards are the law’s attempt to reach the ‘next best’ position tothe wrong not having been committed. For breach of contract, this is thenext best position to the performance having been rendered. The paper forthe symposium at which this article was originally presented was due on 31December 2007. By 1 January 2008 I could no longer keep my promise,but this did not relieve me from my obligation. If I can get a presentablearticle in by 2 January, that is what I must do. Damages are not owed inorder to make good or eradicate losses, any more than is my obligation tomake a late delivery coupled with an apology. Eradicating losses willfrequently be impossible through a monetary award, as most obviouslywhere the plaintiff has lost a limb or suffers terrible emotional distress as aresult of the defendant’s wrong. I cannot eradicate the stress of theorganiser waiting for my article to arrive. Rather damages seek to achievethe closest position to the wrong not having occurred as can be ordered.The damages in substitution for the right which were awarded in Ashby vWhite or The Mediana represent the law seeking to achieve the next bestthing to the right not having been violated in the first place. This may notbe a particularly close approximation to the position which would other-wise have obtained, just as paying money for the loss of a limb will notcause it to grow back, but, given where we are, it is the best which can beordered to be done.

14 J Edelman, ‘Gain Based Damages and Compensation’ in A Burrows and Lord Roger(eds), Mapping the Law: Essays in Memory of Peter Birks (Oxford, Oxford University Press,2006).

15 Above n 13, at 577 (Lord Reid), 580 (Lord Morton); contra The Bodlewell [1907] 1 P286; cf Admiralty Commissioners v SS Valeria [1922] 2 AC 242 (HL).

16 See also The Sanix Ace [1987] 1 Lloyd’s Rep 46 (QBD); The Charlotte [1908] P 206(CA); and The Winkfield [1902] P 42 (CA).

174 Robert Stevens



Another way of demonstrating the importance of substitutive awards iswhere the defendant’s action constitutes two separate wrongs to theplaintiff, which lead to different awards of damages. If an employersummarily fires an employee, the employee will have a claim for breach ofcontract. If, however, the breach also amounts to an act of discrimination,perhaps on the basis of sex or race, the plaintiff will have an alternativecause of action.17 The loss suffered by the plaintiff may not differaccording to whether the dismissal constitutes an act of discrimination ornot. However, where it does so constitute discrimination greater damagesare payable. The violation of this additional right leads to a higher award,despite the absence of any additional loss.

On one view, the mystery is not why the law awards damages basedupon the value of the (contractual) right infringed, but why it goes furtherand allows the plaintiff to recover for his losses over and above the valueof such right. Again, this is because it is the ‘next best’ to the wrong nothaving occurred. Paying money to make good consequential loss may notbe the best that the defendant can do, but will usually be the best whichcan be ordered by a court to be done. For example, when I fail to do whatI have promised, one aspect of the obligation to achieve the ‘next best’ is toapologise. The courts make money awards not because this is the next bestwhich can necessarily be achieved but because this is the next best thingthat they can order to be done. Involuntary apologies are not apologies atall.

All forms of damages awards for all wrongs, including breach ofcontract, seek to place the claimant in the position he or she would havebeen in if the wrong had not occurred, in so far as money can do this.There is no difference in this regard between damages for torts, equitablewrongs or breaches of contract. The underlying primary rights may bedifferent in content, but the aim of the secondary obligation to paydamages for their infringement does not vary. This means that damages forbreach of contract seek to place the claimant, so far as money can do it, inthe position he or she would have been in if the contract had beenperformed. This is the next best position. In rare cases this may beincapable of proof. If, for example, an actor agrees to appear in a film, butin breach refuses to perform resulting in the film’s cancellation, whetherthe film would have been a success may be impossible to demonstrate.18 Itcannot be shown what the position upon performance would have been. Insuch exceptional circumstances damages may be awarded which are basedupon the position that the plaintiff would have been in if the contract hadnever been entered into. Such damages are the ‘next-next-best’ which can

17 See Sex Discrimination Act 1975 (UK); Race Relations Act 1976 (UK).18 See, eg, Anglia Television Ltd v Reed [1972] 1 QB 60 (CA).




be awarded, and are only ordered where the next best position isunknowable.19 However, for the purposes of this article the importantpoint is that the ‘next best’ is not necessarily calculated by reference to losssuffered.

Not all wrongs give rise to an entitlement to substantial damageswithout proof of consequential loss. Deceit, for example, requires proof ofconsequential loss and the only remedy available is one to make good theloss suffered. Lies, per se, are not actionable. Slander, subject to exceptionsand unlike libel, is similar. That breach of contract does not fall into thecategory of wrongs which require the proof of consequential loss beforesubstantial damages will be awarded, is probably most readily demon-strated in the context of the sale of goods.

III. SALE OF GOODS

A. Non-delivery

Where a seller in breach of contract fails to deliver, the buyer is entitled tothe difference between the market value of the goods and the contractprice.20 This rule cannot be displaced by demonstrating that the buyersuffered no loss consequent upon the seller’s breach. This was establishedby the House of Lords in Williams Bros v Ed T Agius Ltd.21 The defendantfailed to deliver a consignment of Norwegian coal which it had agreed tosell to the plaintiff for 16s 3d per ton CIF Genoa (November shipment).The plaintiff had agreed to sub-sell coal of the same quantity anddescription for 19s to X. It was intended that the cargo the plaintiffsbought from the defendant would be used to fulfil the contract with X butthey were under no obligation to do so. The market price at the time ofbreach stood at 23s 6d.

The defendants argued that where the plaintiff has in fact suffered lessthan the difference between the contract price and the market price at thetime of the breach, then he is not entitled to the full market price. Theplaintiffs had not bought in substitute goods at the higher market price andhad no intention to do so, and there was no chance of their being sued for

19 C & P Haulage v Middleton [1983] 1 WLR 1461 (CA).20 Sale of Goods Act 1979, s 53(3). The sale of goods legislation is a codification of the

common law of contract in this context.21 [1914] AC 510 (HL) [Williams Bros]. See also Diamond Cutting Works Federation Ltd

v Triefus & Co Ltd [1956] 1 Lloyd’s Rep 216 (QBD) (Barry J); Mouat v Betts Motors Ltd[1959] AC 71 (PC NZ); Derby Resources AG v Blue Corinth Marine Co Ltd (The AthenianHarmony) [1998] 2 Lloyd’s Rep 410 (QBD) (Colman J). But see Sealace Shipping Co Ltd vOceanvoice Ltd [1991] 1 Lloyd’s Rep 120 (CA) criticised by GH Treitel, ‘Damages forNon-Delivery’ (1991) 107 LQR 364.

176 Robert Stevens



non-delivery by the sub-buyer.22 The House of Lords refused to discountthe damages payable, applying the market value rule.

If, as is commonly thought, contract damages are only awarded tocompensate for actual loss suffered, ignoring the sub-sale contract wronglyover-compensates the plaintiffs.23 What William Bros demonstrates is thatcompensating for loss is not the sole purpose of contract damages.

Applying the market value rule, and ignoring sub-sales, might bethought to be justified by the goal of commercial certainty. The task offixing the appropriate level of damages is made easier by ignoring thesub-sale. However, that this is not the true justification for the rule can bedemonstrated by those cases where the sub-sale is taken into accountbecause the consequential loss the plaintiff suffers as a result of the termsof the sub-sale is higher than would normally have been suffered. Nor-mally, in the case of generic goods for which there is a ready market therewill be no consequential loss, as the buyer will be able to go into themarket and fulfil the sub-sale contract.24 However, this will not always beso. This may be illustrated by another decision of the House of Lords, ReHall (R and H) Ltd and Pim (WH)(Jnr) and Co’s Arbitration,25 whosefacts closely resemble Williams Bros. On 3 November, the defendantagreed to sell to the plaintiff a cargo of Australian wheat CIF UK ports at51s 9d a quarter. On 21 November, the plaintiffs sub-sold wheat of thesame quantity and description to X at 56s 9d. X, in turn, had agreed to sellsuch a cargo to Y at 59s 3d a quarter. The defendant bought a cargo ofwheat on board SS Indianic at 60s a quarter. The defendant securedagreement with all concerned that the sales should each be treated as are-sale of the cargo which was the subject matter of the preceding purchasein the chain. The defendant gave notice appropriating the Indianic cargo toits contract with the plaintiff and that notice was passed down the chain.The defendant sold the Indianic cargo to a third party at 59s 1/2d aquarter. When the cargo arrived, the market price was 53s 9d a quarter.Having sold the cargo, the defendant was unable to deliver the documentscovering the cargo to the plaintiff. The Court of Appeal held that Hall’sdamages were limited to the difference between the market price (53s 9d)

22 This was either because the terms of the sub-sale contract relieved the plaintiffs of theirobligation to deliver when no delivery to them took place, or because X had in turn soldgoods of the same quantity and description back to the defendant, assigning to the defendantits rights against the plaintiff under the contract of sub-sale. In order to rely upon the rightsunder the assigned sub-sale the defendant should have raised a counter-claim.

23 AS Burrows, Remedies for Torts and Breach of Contract, 3rd edn (Oxford, OxfordUniversity Press, 2004) 213; cf SM Waddams, The Law of Damages, 3rd edn (Aurora,Canada Law Book, 1997) 1.1360–1.1440, 1.1930, 1.1940.

24 See Kwei Tek Chao v British Traders & Shippers Ltd [1954] 2 QB 459, 489–90 (DevlinJ).

25 (1928) 33 Com Cas 324 (HL). See also Household Machines Ltd v Cosmos ExportersLtd [1947] KB 217.




and the contract price (51s 9d) at the date of the breach. The plaintiffclaimed the difference between the price under their sub-sale to X (56s 9d)and the contract price (51s 9d).

The House of Lords restored the decision of Rowlatt J that the plaintiffwas entitled to recover the difference between the price at which it hadbought and the price at which it had resold the cargo together with anindemnity for the damages and costs which the plaintiff would have to payto the buyers who had bought from them. Where what is claimed isconsequential loss over and above the market value of the goods thisshould be recoverable unless it is too remote. It was well known to bothparties that it was common practice to resell cargoes whilst afloat. Apartfrom common knowledge, the contract itself showed this. Moreover, thecorrespondence as to the actual appropriation of the vessel was additionalproof, if proof were needed, of the familiarity of the defendant with thepractice of successive re-sales of cargo afloat. The defendant knew as soonas it nominated a cargo that only delivery of that cargo could satisfy thecontract, and it was sufficient to give rise to liability for loss of profit thatthere was an even chance of a sub-sale taking place.

It is not correct therefore that sub-sales are ignored. If the sub-saleincreases the loss suffered as a result of the defendant’s breach, it should berecoverable unless too remote.26 However, where the actual loss suffered islower than the market value because of the sub-sale, this does not reducethe damages payable as the buyer can claim damages based upon his or herright to performance regardless of the consequential loss actually suffered.

The justification for the market value rule in Williams Bros is not,therefore, the promotion of a commercially certain rule fixing damages at aparticular point. Sometimes sub-sales are ignored and sometimes they aretaken into account. Further, the rule is not justified on the basis that ‘whatis sauce for the selling goose is sauce for the buying gander’, as is alsosometimes suggested.27 The fact that sub-sales can be brought into accountby the buyer to increase the damages payable over and above the marketvalue assessment, but cannot be used by the seller by way of reduction, isonly explicable on the basis that sometimes damages are measured by theloss suffered, and sometimes they are not.

The Ontario Law Reform Commission proposed to overturn the com-mon law in cases of non-delivery and confine the claimant to damages foractual loss suffered, reversing Williams Bros.28 If piecemeal reform of this

26 For cases where such loss was too remote: see Williams v Reynolds (1856) 6 B & S 495,122 ER 1278; Aryeh v Lawrence Kosteris & Son Ltd [1967] 1 Lloyd’s Rep 63 (CA);Burgoyne v Murphy [1951] 2 DLR 556 (NBSC AD).

27 Williams Bros, above n 21, at 523 (Lord Dunedin).28 Ontario Law Reform Commission, Report on Sale of Goods (Toronto, Ministry of the

Attorney General, 1979) 502.

178 Robert Stevens



kind is adopted it will create anomalous differences between damages inthe law of sale and in other areas, such as carriage.

B. Defective Goods

Until recently, the position in England in relation to defective goods couldbe confidently stated to be consistent with the law on non-delivery, asillustrated by the leading case of Slater v Hoyle & Smith Ltd.29 In thatcase, the plaintiff buyers bought from the defendant sellers a quantity ofunbleached cotton cloth, some of which they then used, after bleaching, tofulfil a contract of sale to a third party. The cloth supplied was of inferiorquality to that warranted under the contract. The price under the sub-salecontract was paid in full, and no claim was brought by the sub-buyersagainst the plaintiff. The sellers argued that the buyers had suffered no lossas a result of the defect in quality. The Court of Appeal upheld the awardof damages based upon the difference in market value at the time ofdelivery between the defective cloth and cloth conforming to the qualityrequirements of the contract.

Various arguments can be constructed to attempt to show that thebuyers did in fact suffer a loss, but none is convincing. For example, itcould be said that the buyers were not bound to deliver the same goodsunder the sub-sale that they received under the contract of sale. The buyercould have purchased other goods on the market to fulfil the sub-sale.However, in quantifying loss we are interested in what actually did happen:on the facts the buyers did not go out into the market in this way.

Similarly, it might be said that the buyers have paid too much for thegoods, and that this loss is a real loss. However, damages are notquantified by reference to what, if anything, the plaintiff overpaid at thetime of contracting but rather by reference to the value of the contractualright at the time of performance. If, for example, £100 is paid for clothwhich, as delivered, has a value at the time of contracting of £80 or whichthe plaintiff would have been prepared to pay £80 for if he or she hadknown of its quality at time of contracting, damages are not calculated bylooking at the difference between what was paid and what would havebeen paid. If, therefore, the market rises so that at the time of delivery themarket value of the goods as promised is £120, but the value of the goodsdelivered is at that time £90, damages of £30 are payable, not £20.

Unfortunately, Slater was not followed by the Court of Appeal in BenceGraphics International Ltd v Fasson UK Ltd.30 The defendants sold aquantity of vinyl film to the buyer who used it to make decals that were

29 [1920] 2 KB 11 (CA) [Slater].30 [1998] QB 87 (CA) [Bence].




used in the shipping industry to identify bulk containers. It was arequirement of the container trade that the containers would last for fiveyears, so it was stipulated in the contract of sale that the film would lastfor five years, so that what was printed upon it would last for the industrystandard period of time. However, the sellers supplied sub-standard vinylwhich did not survive the contracted-for period. The buyers receivedextensive complaints from end-users. This resulted in only one minor claimthat the buyers settled and for which they were reimbursed by the sellers.The trial judge awarded damages by reference to the market value of whatwas delivered. The Court of Appeal by a majority allowed the appeal,awarding a small sum with respect to that film which was unused andremitting the case for the assessment of consequential loss, based upon theplaintiff buyer’s liability to the ultimate end-users of the film.

The case has been convincingly criticised,31 and even its supporters concedethat the reasoning of the majority is unconvincing. First, the majority treatedthe question in issue as one of remoteness. This was a mistake. Wheredamages are claimed for the difference in value between what was promisedand what was delivered, no issue of remoteness arises.32 Only where the claimis for consequential loss does the issue of remoteness arise. Second, themajority failed to provide any convincing distinction between Bence andSlater, the latter being a unanimous decision of Bankes, Warrington andScrutton LJJ. Otton LJ sought to distinguish Bence on two narrow bases: first,that the goods sold on in Slater were substantially the same, whereas in Bencethey had been substantially processed; and second, that in Slater the seller hadnot known of the contemplated re-sale whilst in Bence they had. The firstground is both unconvincing on the facts, and of no relevance to the basis ofthe market rule. The second is also questionable as a matter of fact, difficult tonormatively understand, and even if accepted, would indicate exactly theopposite results: substantial damages in Bence and no recovery in Slater. AuldLJ was bolder still, failing to distinguish Slater but refusing to follow it.

Bence is wrong in reasoning and result. It is flatly inconsistent with priorCourt of Appeal authority and inconsistent with the rationale behind therule in cases of non-delivery established by the House of Lords in WilliamsBros. In Canada Bence can be safely ignored as, on facts materiallyidentical to Slater, the Supreme Court in Bainton v John Hallam Ltdreached the same result as had the English Court of Appeal five monthsearlier.33

31 GH Treitel, ‘Damages for Breach of Warranty of Quality’ (1997) 113 LQR 188; CHawes, ‘Damages for Defective Goods’ (2005) 121 LQR 389; MG Bridge, ‘Defective Goodsand Sub-sales’ [1998] Journal of Business Law 259.

32 Treitel, ibid, at 190.33 [1920] 60 SCR 325, 54 DLR 537. Slater was not cited, presumably because it was not

yet reported.

180 Robert Stevens



Superficially difficult to reconcile with Slater is the earlier decision of theHouse of Lords in British Westinghouse Electric and Manufacturing CoLtd v Underground Electric Railways Co of London Ltd.34 In 1902 thedefendants agreed to sell eight steam turbines for electricity generation tothe plaintiff railway company for £250,000. The turbines proved to bedefective. The plaintiff sought damages of upwards of £280,000 for lossesthat they estimated would be caused by the excessive coal consumptionover the life of the machines. Alternatively, they claimed the cost ofinstalling eight new turbines, with superior kilowatt capacity, which theyhad purchased in 1908 when the seller’s machines proved insufficient.They estimated the cost of these machines to be £78,186. Even if theseller’s machines had complied with the conditions of the contract, itwould still have been to the advantage of the buyers (at their own cost) tohave replaced the machines supplied with the new machines as soon as thelatter became available on the market.

It should be apparent that the claim for damages as framed was one forconsequential loss. These losses were not in fact suffered because ‘thesuperiority of the [replacement] machines and of their efficiency inreducing working expenses was in point of fact such that all loss wasextinguished, and that actually the respondents made a profit by the coursethey took’.35 The claim was not one for the difference in market valuebetween the machines promised and those delivered. It was not a claimbased upon the value of the right to performance. Indeed such a claimcould not have been brought in addition to that which was asserted. Theplaintiff had recovered substantial damages for the losses incurred becausethe machines were defective during the years up until their replacement. Aplaintiff cannot recover both the difference in value between what it waspromised and what it received, and the expense it in fact incurs in makinggood the defective performance. Recovering the former means that thelatter loss is, to that extent, not incurred. If my wife had insisted upon anew platinum ring I could not have claimed by way of damages both thedifference in value between the ring promised and delivered, and inaddition the cost of replacing it. If I recover the former, to that extent, I donot suffer the latter. Recovery under one head reduces the damagesrecoverable under the other.

Where, however, a claim for damages based upon the value of thecontractual right is brought, the issue of mitigation which arose in BritishWestinghouse does not arise. Questions of mitigation, like remoteness,only arise in relation to consequential losses. So, British Westinghouse may

34 [1912] AC 673 (HL) [British Westinghouse]. See also Erie County Natural Gas & FuelCo v Carroll [1911] AC 105 (PC Can); Cockburn v Trusts & Guarantee Co (1917) 55 SCR264, 37 DLR 701.

35 British Westinghouse, ibid, at 688.




be usefully contrasted with the subsequent decisions of the Privy Council inJamal v Moolla Dawood Sons & Co36 and the Court of Appeal inCampbell Mostyn (Provisions) Ltd v Barnett Trading Co.37 Consequentiallosses are assessed at the time of judgment, taking into account eventswhich have happened subsequent to the wrong that increase or decreasethe loss which would have been expected to have been suffered. This wasthe approach in British Westinghouse. By contrast, in assessing damageswhich are substitutive for the right, the valuation of the right takes place atthe time of infringement.38 In cases of non-delivery, or the delivery ofdefective goods, this is at the time of delivery. Subsequent events areirrelevant as the court’s task is not to calculate what actual loss has beensuffered. The correct question is: what is the difference in value at the timeof breach between the performance promised and the performancereceived?

In Jamal the contract concerned the sale of a quantity of shares for185,000 rupees, with delivery to take place some months later on 30December 1911. The shares fell substantially in value and the defendantbuyer refused to complete. Subsequent to the buyer’s repudiation, the sellerbegan to sell the shares elsewhere. The seller managed to sell at pricessignificantly higher than the market price at the time of breach, therebyavoiding the loss he would otherwise have suffered. The defendant buyerargued that this should be taken into account in quantifying damages. ThePrivy Council correctly rejected this contention, assessing damages byreference to the value of the right to performance at the time of breach.

Precisely the same is the decision of the Court of Appeal in CampbellMostyn. The defendant buyer wrongfully rejected 350 tins of SouthAfrican ham which they had agreed to purchase. Two weeks after thebreach, the government announced restrictions on ham imports fromContinental Europe. This action increased the price at which the plaintiffswere able to re-sell and resulted in the sellers suffering no loss as a result ofthe buyer’s breach. According to the court, the plaintiff’s successfulmitigation of any loss they might have suffered did not reduce the damagespayable, these being assessed by reference to the market price at the time ofbreach. Again, the actual loss suffered at the time of judgment was ignoredbecause the damages here, unlike in British Westinghouse, did not seek tocompensate for the actual loss suffered.39

36 [1916] 1 AC 175 (PC) [Jamal]. Viscount Haldane LC who delivered the speech for thecourt in British Westinghouse was also a member of the panel in Jamal. For a sale of goodsexample: see Jones v Just (1868) LR 3 QB 197.

37 [1954] 1 Lloyd’s Rep 65 (CA) [Campbell Mostyn].38 See generally Stevens, above n 5, at 69–70.39 An unsatisfactory exception is made when, after a buyer has wrongfully rejected goods

under a contract, and a new agreement is entered into by the same parties for the sale andpurchase of the same goods at a reduced price. Where the buyer sues the seller for damages

182 Robert Stevens



C. Late Delivery

In Slater, Scrutton LJ was strongly critical of the earlier decision of thePrivy Council in Wertheim v Chicoutimi Pulp Co.40 However on the basisof the thesis put forward here both are readily reconcilable. The contractwas for the sale of 3,000 tons of Canadian moist pulpwood FOBChicoutimi for delivery by November 1900. The buyers had alreadycontracted to sell 2,000 tons of pulpwood, and subsequently madecontracts which accounted for the balance of the FOB quantity. Thepulpwood was not in fact shipped until June 1901. The buyers were able toperform the sub-sale contracts despite the delay.

The market price prevailing had the goods been delivered on time was70s, but it had fallen back to 42s 6d by the time delivery was actuallymade. The plaintiff sought to recover the difference in market value: 27s6d. The sub-sale price of the goods was 65s and the plaintiff was able toperform these contracts despite the delay. In light of these facts, the Houseof Lords refused to make an award for loss which was not actuallysuffered.

What is the distinction between the late delivery in Wertheim and thenon-delivery in Williams Bros? The diligent reader of the speeches of theHouse of Lords and the advice of the Privy Council will not find asatisfactory explanation. Again, the difference can be explained by recog-nising that damages may be awarded as a substitute for the right infringed.What was the difference in value between what was promised and whatwas received at the time of breach? More precisely, what was the differencein value between a delivery on 1 November and a June delivery for aperson without knowledge of the future events, at the start of November1900? At that time the market fall had not yet occurred, and so thedifference would either have been nil or very small. If the plaintiff wishedto claim damages for the loss suffered as a result of the market fallsubsequent to the breach, this loss would actually have had to have beensuffered which it was not.

One mystery does remain, however. The Privy Council did uphold anaward of damages based upon the difference between the market priceprevailing at the due date of delivery (70s) and the sub-sale price (65s).How this award of 5s was arrived at is a mystery as it neither reflects thevalue of the right nor the loss suffered. The best explanation may be thatthe plaintiff was entitled to some damages because of the violation of his

under the original contract, account can be taken of the profit made by the buyer on thesubsequent contract, provided that it is found as a fact that the subsequent contract is part ofa continuous dealing between the parties: see Pagnan & Fratelli v Corbisa IndustrialAgropacuaria Ltda [1970] 1 WLR 1306 (CA).

40 [1911] AC 301 (PC) [Wertheim].




right to timely delivery, and that the defendant had conceded that this sumwas an appropriate level of damages.

IV. CARRIAGE OF GOODS

The position in relation to the sale of goods is also found in relation totheir carriage. Rodocanachi Sons & Co v Milburn Bros concerned anaction for non-delivery under a voyage charter of a cargo of cotton shippedat Alexandria on account of the charterer and bound for the UnitedKingdom.41 Owing to the Master’s negligence the cargo was lost. Thecharterer had sold on the cargo on a ‘to arrive’ basis, relieving the chartererof any obligation should the goods fail to arrive. The sub-sale was at aprice considerably below the market price when the goods should havearrived in the United Kingdom. The contention of the defendant—namely,that the plaintiff ought not to be placed in a better financial position thanwould have obtained if the contract had not been broken—was rejected bythe Court of Appeal. The value of the goods at the time and place ofdelivery was recoverable ‘independently of any circumstances peculiar tothe plaintiff, and so independently of any contract made by him for the saleof the goods’.42

Similar is the important decision of the House of Lords in Ströms BruksAktie Bolag v Hutchison.43 The claim was for non-delivery under a voyagecharter from Sweden to Cardiff of a cargo of wood pulp to be lifted in twoshipments, the second in August–September at the defendant carrier’soption. The contract was made to enable the plaintiff to fulfil a contract ofsale which provided for the delivery of the cargo in Cardiff. The terms ofthe contract of sale and the contract of carriage did not therefore align, theplaintiff charterer’s mistake appearing to have arisen from the fact thatalthough the contract was described as being a CIF sale, it was by its termsone for the delivery of the goods. The defendant argued that even if theyhad performed their contract to the letter (that is, if the cargo had beenlifted at the end of September) this would not have enabled the plaintiff toperform the contract of sale, the benefit of which would have been lost inany event. They should not, therefore, be liable for loss which would havebeen suffered anyway. The House of Lords rejected this argument. Therewas no need to prove ‘special’ damage. The contract of sale was simply the‘best evidence possible’ of the general damages suffered by the plaintiff.44

41 (1887) 18 QBD 67 (CA) [Rodocanachi Sons].42 Ibid, at 77 (Lord Esher MR).43 [1905] AC 515 (HL).44 Ibid, at 526 (Lord Macnaghten). See also Biggin v Permanite [1951] 1 KB 422, 438

(Devlin J); Eldon Weiss Home Construction Ltd v Clark (1982) 39 OR (2d) 129 (Co Ct) 133.

184 Robert Stevens



Put another way, it was the best available evidence of the value of the rightto have the goods delivered in Cardiff at that date.

The decision of the Court of Appeal in White Arrow Express Ltd vLamey’s Distribution Ltd concerns the provision of a completed, ifdefective, carriage.45 The Court of Appeal concluded that where theclaimant has bargained for a deluxe delivery service and has received onlya standard service, he or she is entitled to the difference in market valuebetween what was received and what was bargained for, although theclaimant failed to establish any such difference on the facts. Sir ThomasBingham MR stated:

It is … obvious that in the ordinary way a party who contracts and pays for asuperior service or superior goods and receives a substantially inferior service orinferior goods has suffered loss. If A hires and pays in advance for a 4-doorsaloon at £200 per day and receives delivery of a 2-door saloon available for£100 per day, he has suffered loss. If B orders and pays in advance for a 5-coursemeal costing £50 and is served a 3-course meal costing £30, he has suffered loss.If C agrees and pays in advance to be taught the violin by a world famouscelebrity at £500 per hour, and is in the event taught by a musical nonentitywhose charging rate is £25 per hour, he has suffered loss.46

Whilst Sir Thomas Bingham MR describes the recipient of the service ashaving suffered a loss, this is not a consequential loss. In the examplesgiven, the claimant is not, or at least is not necessarily, left in a factuallyworse position after performance than if he or she had received thebargained-for service. The ‘loss’ constitutes failing to receive performanceof the bargained-for quality. His approach is correct and ought to beapplied to all contracts for the provision of services just as it does tocontracts for the sale of goods.47 So if E contracts with F for the provisionof an ordinary service, and receives an inferior one, E should be entitled toa claim for damages, just as should the person who has contracted for asuperior version.

A Canadian decision which is apparently out of line with these principlesis the decision of the Federal Court of Appeal in Redpath Industries Ltd vThe Cisco.48 The defendant carrier carelessly damaged the plaintiff charter-er’s cargo of sugar by seawater. The sugar as damaged was only sellable asanimal feed for $53,000, whilst undamaged it would have a value of$280,000. The plaintiff was a sugar refiner who was able to refine smallamounts of wet sugar with large amounts of dry so that it was sellable as

45 (1996) 15 Tr LR 69.46 Ibid, at 73.47 See also Attorney General v Blake [2001] 1 AC 268 (HL) 286 (Lord Nicholls) [A-G v

Blake].48 (1994) 110 DLR (4th) 583 (CA) leave to appeal to the SCC refused 116 DLR (4th) vii.




undamaged. This refining cost about $50,000. Damages were confined tothe lower sum, the court treating the claim as one for a loss which had infact been mitigated.

V. SALE OF LAND

Unsurprisingly, the position found in relation to contracts for the sale ofgoods is reflected in claims for breach in relation to contracts for the saleof land.49 However, as land is not generic, specific performance is com-monly available, leading to possible differences in relation to the timing ofassessment of the value of the right.

Where specific performance or an injunction is awarded the court is notordering a remedy for breach of contract but rather is compelling theperformance of the primary contractual right. If, for example, a seller ofland repudiates the contract, the buyer may reject his or her repudiationand seek performance. A repudiation which is not accepted is a thing writin water,50 which does not constitute a breach. Specific performance maybe awarded even though there is no actual breach of contract.

Where damages are awarded ‘in lieu of specific performance’ or ‘in lieuof an injunction’ such damages are also substitutive. They are awarded inlieu of the right to performance. Again, proof of consequential loss isunnecessary.

This may be illustrated by the decision of the Supreme Court of Canadain Semelhago v Paramadevan.51 The plaintiff agreed to buy a house fromthe defendant in the Toronto area for $205,000. The defendant sellerreneged on the deal, and conveyed the property to a third party. The valueof the property rose between the time of repudiation and trial to $325,000.However, the plaintiff kept his own house which rose in value during thesame period from $190,000 to $300,000. The defendant argued that incalculating damages, the court should take into account the fact that hadthe contract been performed the plaintiff would not have acquired thebenefit of the increase in value of his old property because he would havehad to have sold it.

When damages are awarded in lieu of specific performance the correctmoment for assessing the value of the right is the time of judgment. It is theright to performance at that time for which damages are awarded assubstitute. This may be contrasted with a claim based upon a pre-trial

49 See Brading v McNeill & Co Ltd [1946] Ch 145.50 See Howard v Pickford Tool Co Ltd [1951] 1 KB 417 (CA) 421 (Asquith LJ).51 [1996] 2 SCR 415, 136 DLR (4th) 1. See L Smith, ‘Understanding Specific Performance’

in N Cohen and E McKendrick (eds), Comparative Remedies for Breach of Contract (Oxford,Hart Publishing, 2005) 221, 228–30.

186 Robert Stevens



breach, where the right is valued at the time of breach.52 The award inSemelhago was not assessed according to the gain the defendant madefrom the breach—an issue which the court did not examine. Indeed thegain made by the increase in value was made by the third party, not thedefendant vendor. Rather the award monetarises the right to performance,even though this leads to an award over and above any loss suffered by theplaintiff.

It may be objected that awarding damages calculated by reference to thevalue of performance at time of trial is unfair to defendants as plaintiffs areallowed to speculate, without risk, on market movements. If the housegoes up in value after the date upon which performance was due, plaintiffscan bring a claim and capture this gain for themselves, if the marketdeclines they can leave the house with the defendants. However, there is noinjustice to defendants once it is accepted that damages in lieu of specificperformance are an indulgence: they are afforded where they do not wishto perform. If defendants wish to forestall speculation on the market theyare always able to do so by performing their promises. Even as late as thedate of trial, defendants can opt to perform rather than pay damages.

Lord Cairns’ Act allows damages assessed according to the value of theright at the time of judgment to be awarded.53 It is this ability which itadds to the court’s armoury. It is unclear what purpose those who considerdamages to be always compensatory think Lord Cairns’ Act is serving.

VI. COVENANTS TO REPAIR

In Joyner v Weeks a tenant was in breach of a covenant of repair.54 Thelandlord suffered no loss because he had entered into another lease underwhich the subsequent tenant agreed to affect repairs. If damages were onlyawarded for consequential loss, the landlord’s claim ought to have failed ashe had mitigated any loss after the tenant’s breach. The claim forsubstantial damages succeeded. As the Court of Appeal stated many ‘casesmay be put in which it is plainly immaterial that at the commencement ofan action for a breach of contract the plaintiff is in fact no worse off thanhe would have been if the contract had been performed’.55 The value of thelandlord’s right to the work was quantified according to the difference invalue between the repaired and unrepaired premises, capped by the cost ofdoing the repairs.

52 See Wroth v Tyler [1974] Ch 30 (Megarry J) but see Johnson v Agnew [1980] AC 367(HL).

53 Chancery Amendment Act 1853, s 2 (now contained in Supreme Court Act 1981, s 50).54 [1891] 2 QB 31 (CA); Smiley v Townshend [1950] 2 KB 311 (CA); Haviland v Long

[1952] 2 QB 80 (CA).55 Joyner v Weeks, ibid, at 33.




VII. BUILDING WORK

In Alfred McAlpine Construction Ltd v Panatown Ltd, Lord Goff gave thefollowing example illustrating the same principle in the context of buildingwork:

[A] wealthy man who lives in a village decides to carry out at his own expensemajor repairs to, or renovation or even reconstruction of, the village hall, andhimself enters into a contract with a local builder to carry out the work to theexisting building which belongs to another, for example to trustees, or to theparish council. Nobody in such circumstances would imagine that there could beany legal obstacle in the way of the charitable donor enforcing the contractagainst the builder by recovering damages from him if he failed to perform hisobligations under the building contract, for example because his work failed tocomply with the contract specification.56

As Lord Goff makes clear the recovery of substantial damages is notdependent upon the plaintiff incurring any actual loss: ‘Is it really to besuggested that his action will fail, because he does not own the hall, andbecause he has not incurred the expense of himself employing anotherbuilder to do the remedial work?’57

That Lord Goff’s approach is correct is apparent from the analysis givenin the previous sections of this article. However, two members of the court,Lord Jauncey and Lord Clyde, rejected his approach and McAlpine vPanatown has left the law in a state of uncertainty.

The facts of the case were as follows. Panatown employed McAlpine tobuild an office block and multi-storey car park on a site in Cambridge.Panatown alleged that the building was seriously defective. The site wasnot, however, owned by Panatown but by its holding company UnexInvestment Properties Ltd (‘UIPL’). On the same day that the constructioncontract was entered into, McAlpine signed a duty of care deed in favourof UIPL undertaking to take reasonable care and skill in carrying out thebuilding contract.

Panatown sought to recover damages against McAlpine on two grounds.First, they argued that they were entitled to recover damages on behalf ofthe third party (the ‘narrow ground’). This ground need not be consideredfurther. Second, they argued that they were entitled to damages in theirown right because they had not received what they were promised (‘thebroad ground’). The broad ground was accepted by Lord Goff and LordMillett, who dissented, but rejected by Lord Jauncey and Lord Clyde. LordBrowne-Wilkinson who had expressed sympathy for the broad ground inLinden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd58 and conceded

56 [2001] 1 AC 518 (HL) 547 [McAlpine v Panatown].57 Ibid, at 548.58 [1994] 1 AC 85 (HL) 112.

188 Robert Stevens



that examination by academic writers had discovered no serious difficultieswith it, was not prepared to allow it to succeed where the third party hada right of enforcement in his or her own right.

In the example given by Lord Goff, it is submitted that there would, onany view, be a loss if, after the work had been done, the philanthropistreasonably employed a second builder to remedy the defective work or atleast was likely to do so.59 Can the promisee recover substantial damageseven where this is not so? Lord Goff and Lord Millett would allowrecovery in such circumstances. In the example given by Lord Goff, thethird party would have no contractual right against the promisor, and soLord Browne-Wilkinson may well also have been prepared to allowrecovery.60 Whether Lord Browne-Wilkinson was right to refuse to allow aclaim for substantial damages in McAlpine v Panatown itself depends uponthe correct construction of the agreement. Was what was bargained for anoffice block and multi-storey car park, or was what was bargained for sucha building or a claim for damages by UIPL? Did the right to claim underthe deed exclude the ability to claim damages on any other basis, just aswould a liquidated damages clause? If the latter construction is adopted,the refusal of Panatown’s claim for damages seems unobjectionable.However, on the more plausible construction of the parties’ deal, the dutyof care deed was required not in order to give UIPL a claim for damages inits own right, but rather for the benefit of subsequent purchasers of thepremises.61 On the better construction of the bargain, the claim fordamages in McAlpine v Panatown ought to have succeeded.

Although at certain points Lord Goff (and Lord Millett) seem to indicatethat the appropriate method for quantifying the claim for damages is thecost of repairs, this would be a mistake. Repair costs are a measure of loss.Where not actually incurred, such costs are only of relevance in valuing theright to performance as a matter of evidence. That damages are notawarded for actual expenses incurred is apparent from Lord Goff’streatment of damages for delay. The employer has ‘a contractual right tothe performance by the contractor of his obligation as to time, as much ashe has to his performance of the work to the contractual specification’.62

This right can be

measured objectively in financial terms with reference to the anticipated profit-ability of the development; and this can provide an appropriate yardstick for

59 McAlpine v Panatown, above n 56, at 533 (Lord Clyde), 574 (Lord Jauncey); Radfordv De Froberville [1977] 1 WLR 1262 (Ch D).

60 Cf Woodar Investment Development Ltd v Wimpey Construction UK Ltd [1980] 1WLR 277 (HL) 300 (Lord Scarman); Bovis Lend Lease Ltd v RD Fire Protection Ltd(unreported, QBD, 6 February 2003) (Thornton J, QC).

61 I Wallace QC, ‘Third Party Damage: No Legal Black Hole?’ (1999) 115 LQR 394;McAlpine v Panatown, above n 56, at 593 (Lord Millett).

62 McAlpine v Panatown, ibid, at 554–5.




measuring the estimated damages for delay in the performance for which theemployer has contracted, even where the development was to be carried out on asite belonging to another person.63

Furthermore, to award the full difference in value between the building aspromised and as built, may be incorrect. The plaintiff has a right to thework; the increase in value of the building by the work’s completion maybe greater than this because of the site’s potential for improvement if thework is completed. Damages should therefore be capped at the differencein value between the work promised and the work performed.

The limits of the analysis presented here may be illustrated by consider-ing the case of the disappointed legatees. In White v Jones a testatorchanged his will after a quarrel with his daughters so as to leave themnothing.64 After reconciliation, he instructed his solicitor to draw up a newwill leaving them £9,000 each. The solicitor negligently failed to changethe will, resulting in the daughters inheriting nothing when he died.Although the estate of the testator should have a claim for substantialdamages (since the bargained-for services had not been provided), thisclaim would not be calculated by reference to the sum that the legatees hadfailed to inherit—a sum which may be vastly greater than the value of theservice bargained for. If a claim to the benefit that has not been conferredupon the legatees is to be allowed, it must be done on a different basis,either by allowing the estate to claim this sum on their behalf, or byallowing the legatees a claim in their own right as an exception to thedoctrine of privity of contract.65

McAlpine v Panatown may also be usefully contrasted with the earlierdecision of the House of Lords in Ruxley Electronics and Construction Ltdv Forsyth.66 In that case, the plaintiff contracted for the construction of aswimming pool with a depth of 7 feet 6 inches. This was deeper than theindustry standard because the plaintiff was a tall man who did not like tosee the bottom of the pool when he dived in. The pool, however, was builtto a depth of 6 feet 9 inches which was perfectly safe and usable, but notwhat the plaintiff had bargained for. The cost of increasing the depth of thepool was £21,460. The difference in market value between what waspromised and what was received was zero.

Again we need to distinguish between the value of the contractual rightinfringed and loss consequent upon the infringement of that right. Costs of

63 Ibid.64 [1995] 2 AC 207 (HL).65 See Law Commission, ‘Privity of Contract: Contracts for the Benefit of Third Parties’

(Law Comm No 242, 1996) at [2.14]; P Benson, ‘Should White v Jones Represent CanadianLaw: A Return to First Principles’ in J Neyers, E Chamberlain and S Pitel (eds), EmergingIssues in Tort Law (Oxford, Hart Publishing, 2007) 141; Stevens, above n 5, at 176–82.

66 [1996] AC 344 (HL) [Ruxley].

190 Robert Stevens



cure represent the expenditure which has or will be incurred as a result ofthe defendant’s breach. This is quantified at the time of trial. It is not aquantification of the value of the right to performance at the time ofbreach.67 Its only relevance to the valuation of the right at the time ofbreach is evidential. Where there is no ready market for the goods orservices bargained for, it may provide guidance as to whether the contractprice accurately reflected the right bargained for. However, the costs expost of curing a breach of contract may bear no relation whatsoever to thevalue of the contractual right itself ex ante, as is demonstrated by Ruxley.

Claims for consequential loss, as in Ruxley, are subject to limitationswhich do not apply to claims to the value of the right. If I was summarilyfired by my current employer, I would have a number of options open tome. One of those options would be to spend the rest of my life drinkingbeer in my local public house. If I decided to do so rather than choosing tofind employment elsewhere, which employment would almost certainly bebetter paid than the derisory sum I currently earn, I could not recoverdamages representing my lost salary. Where a plaintiff unreasonablychooses to increase the loss suffered as a result of a wrong, he or she isresponsible for this loss, not the defendant. The primary responsibility forthe welfare of adults lies upon themselves, and this does not change wherewe are the victims of a wrong, whether it is a breach of contract or anyother. There is no ‘duty’ to mitigate, but losses which ought to have beenmitigated cannot be recovered.

Where therefore, as in Ruxley, the decision to incur the costs of cure isunreasonable, it cannot be recovered, as the House of Lords correctly held.The issue before the court was simply one of mitigation of loss.68 Wecannot, as of right, recover all losses which we would not have suffered butfor the defendant’s wrong. By contrast, where there is breach the differencein value between what has been promised and what has been received maybe recovered as of right, regardless of reasonableness.

The House of Lords did uphold the award to the defendant of £2,500.Lord Mustill did so on the basis that ‘the value of the promise to thepromisee exceeds the financial enhancement of his position which fullperformance will secure’.69 Lord Lloyd, by contrast, saw such damages asawarded to make good the plaintiff’s consequential loss of pleasure in nothaving the pool he desired.70 On the facts, the true basis for the awardmade no difference to the result, but the plaintiff ought to be able torecover even where he cannot show that his pleasure is any less as a result

67 Contra Webb, above n 2.68 Burrows, above n 23, at 222.69 Ruxley, above n 66, at 360. See also Farley v Skinner [2002] 2 AC 732 (HL) (Lord

Scott); McAlpine v Panatown, above n 56 (Lord Millett).70 Ruxley, ibid, at 373–5.




of what he received. If, for example, the plaintiff had died shortly aftercompletion of the work his estate should still have a claim for £2,500, eventhough he would not have had the pleasure of using the deeper pool in anyevent. The value of performance is not always measured in terms of themarket difference between what was promised and what was received. Thevalue to the individual claimant, particularly in consumer contracts, mayexceed this, as Lord Mustill contemplated.

VIII. BREACH OF COVENANT NOT TO BUILD

In the light of the forgoing analysis, the result in Wrotham Park v ParksideHomes is readily explicable, but does not have anything to do with a losssuffered by the plaintiff or a gain made by the defendant.71 Land belongingto the defendant was subject to a restrictive covenant registered as a landcharge forbidding development except with approval of the owner fromtime to time of the Wotham Park Estate. The defendant in breach of thecovenant built 14 houses and a road on its land. The plaintiff warned thedefendant of its right as soon as the work commenced, and sought aninjunction restraining the work and seeking the demolition of the buildingwhich had been done. The defendant was (incorrectly) advised that thecovenant was unenforceable and continued work. At trial, Brightman Jrefused to award the injunction, but did order damages, calculatedaccording to a reasonable price for the claimant to have released its right.

Just as in Semelhago, the damages were in lieu of specific relief, and sowere awarded in place of, or in substitution for, the right. Here that rightwas not contractual: there was no privity of contract between the land-owners. Rather, the plaintiff had a right restraining the liberty to use theneighbouring land, a right which would persist against any holder of theright to that land. As there was no ready market for quantifying this right,Brightman J calculated this value as 5 per cent of the £50,000 profit madeby the defendant from the work: £2,500.

There was no loss suffered. The value of the plaintiff’s land was notreduced by the defendant’s building. In some cases it might be argued thatthe relevant loss is the lost opportunity to bargain for the release of thecovenant,72 but as Andrew Burrows pointed out 20 years ago, this will notbe realistic where the defendant would not have agreed to any such release,as was the case in Wrotham Park.73

71 [1974] 1 WLR 798 (Ch D).72 R Sharpe and S Waddams, ‘Damages for Lost Opportunity to Bargain’ (1982) 2 OJLS

290.73 Burrows, above n 23, at 275.

192 Robert Stevens



It does not assist to describe the claim as being made in order tocompensate for an ‘objective’ or ‘normative’ loss. All that such labels showis that the ‘loss’ is not actual or real but rather fictional or deemed. Unlessthe claimant is, as things have turned out, in some way factually worse off(although not necessarily economically) he has suffered no loss. Worse,such expansive use of ‘loss’ is dangerously loose as it misleads us intothinking that damages, which are substitutive for the right, are compensa-tory and are therefore subject to the same principles as apply to damageswhich are awarded to make good consequential losses. As we have seen,however, the rules on remoteness, mitigation, incidental benefits, timing ofassessment and quantification are all different for the two heads ofdamage. We should give different labels to different things, especiallywhere consequences turn upon the difference.

Even worse is the attempt to classify the claim as gain-based. First, theclaim is not quantified according to the actual gain made by the defendant,as is illustrated by Wrotham Park itself. The actual profits or expensessaved do not form the basis of the award, save in so far as they evidencethe value of the right. Second, in a truly gain-based award the defendant’soverall position as a result of the wrong at the time of trial would beexamined. Losses which are suffered as a result of the wrong would becapable of being offset against the gains made. In fact, no such offsetting ispossible, so that a claim for substantial damages would succeed even wheredefendants make overall losses from the wrong.74 Third, describing theclaim as one to the ‘normative’ or ‘objective’ gain is as fictitious asdescribing the claim as one to the ‘normative’ or ‘objective’ loss. As amatter of language, it is possible to describe the infringement of a right as,in itself, a ‘loss’, although as a matter of law this conflation of injuria anddamnum is always unhelpful. To describe the infringement as per se a gainto the defendant is inaccurate, even as a matter of language. The right isnot, as has been claimed,75 transferred to the defendant. The right itself hasnot been ‘gained’, it is still vested in the plaintiff.76 Fourth, the popularityof trying to classify all claims for damages which are not punitive (andhence considered anomalous) as either based upon the plaintiff’s loss or thedefendant’s gain has resulted in the dismissal as incorrect all cases whichcannot plausibly be so classified. So, for example, Burrows considers TheMediana, William Bros, Rodocanachi Sons, Slater, Campbell Mostyn,

74 Inverugie Investments Ltd v Hackett [1995] 1 WLR 713 (PC); See Stevens, above n 5, at79–84.

75 J Edelman, Gain-Based Damages: Contract, Tort, Equity and Intellectual Property(Oxford, Hart Publishing, 2002) 117–18.

76 C Rotherham, ‘The Conceptual Structure of Restitution for Wrongs’ [2007] CLJ 172;Stevens, above n 5, at 80–81.




Jamal, Joyner v Weeks, and Semelhago v Paramadevan all to be wronglydecided, and the minority views of Lords Jauncey and Clyde in McAlpine vPanatown to be preferable.

Fifthly, and most dangerously, if cases such as Wrotham Park aremisclassified as gain-based it is a short step from there to saying that whenthe defendant has made an actual gain from the wrong committed, heshould be required to disgorge it to the plaintiff. However, once therights-based analysis defended here is accepted, the category of gain-baseddamages is rendered otiose save where the actual gain made by thedefendant is greater both than the value of the right infringed and theconsequential loss suffered. It is difficult to identify any cases, prior to thedecision of the House of Lords in Attorney General v Blake,77 whichpermitted such recovery for breach of a contract, or indeed a tort.78

Analogies with the liability of fiduciaries to account for profits does notassist. It is, at the lowest, arguable that such liability is better classified as aprimary rather than a secondary response to wrongdoing. It is one of theprimary duties of fiduciaries that they are obliged to account for the profitsthey make in that capacity. Part of the duty to subordinate your interests tothose of someone else is that you must account for profits you make out ofthe relationship.79 Alternatively, in some cases the duty to account forprofits may be seen as substitutive for the right of the beneficiary whichhas been infringed.

Lord Nicholls, in delivering the majority speech in Blake, describedWrotham Park as a ‘solitary beacon’80 permitting the award of gain-baseddamages for breach of contract. As the claim in Wrotham Park was notbased upon the defendant’s gain, nor indeed was it for breach of contract,it is difficult to resist the conclusion that the beacon caused their Lordshipsto hit the rocks, as Lord Hobhouse forcefully pointed out in his dissent.

IX. NOMINAL DAMAGES

It has been objected that the account of the law given above leaves noroom for the award of nominal damages.81 This is not so.

First, the right infringed may be valueless. If, for example, I wassummarily fired from my job without justification, only nominal damages

77 A-G v Blake, above n 47.78 On the absence of genuinely gain based awards for torts, see Stevens, above n 5, at

83–4.79 P Millett, ‘Equity’s Place in the Law of Commerce’ (1998) 114 LQR 214, 225–7.80 A-G v Blake, above n 47, at 283.81 See A Burrows, ‘Are “Damages on the Wrotham Park Basis” Compensatory, Restitu-

tionary, or Neither?’ in R Cunnington and D Saidov (eds), Contract Damages: Domestic andInternational Perspectives (Oxford, Hart Publishing, 2008).

194 Robert Stevens



would be available to me at common law.82 Although I have suffered awrong, my right to contractual counter-performance from my employer isessentially valueless. My right to be paid my salary is conditional. In orderto earn it, I must work. The work I provide, in terms of teaching, research,supervision and administration is objectively worth far more than thepittance I am paid, as all but the most obtuse observer would accept.Unless I could prove consequential loss, no substantial damages would bepayable as my right to be paid at the end of each month is worth less thanthe work I must do each month in order to get it.

Second, a valuable right may be infringed in an insignificant way. If youscratch my Rolls Royce you obviously need not pay me the full value of thecar. What is quantified is the infringement of the right. If the right isvalueless or the infringement is notional only nominal damages arepayable.

X. THE GOLDEN VICTORY

The decision of the House of Lords in Golden Strait Corporation v NipponYusen Kubishika Kaisha (The Golden Victory)83 may at first sight appearto be inconsistent with the law as I have so far described it.84 Indeed, thespeeches for the majority emphasise that damages for breach of contractare compensatory for actual loss suffered.85 In fact this decision nicelyillustrates the difference between damages in relation to performancewhich has accrued, and those in respect of future performance.

The defendants chartered the plaintiff owner’s vessel under a timecharter. In repudiation of the contract, they redelivered the vessel when thecontract still had four years left to run. The owners accepted the repudia-tion and claimed damages, calculated by reference to the entire four yearsthat the contract still had outstanding. The defendants sought to reduce thedamage payable on the basis that 14 months after the repudiation wasaccepted, the second Gulf War broke out, which would have given bothparties the option to lawfully terminate the contract, an option that thecharterers would have exercised at that point. The defendants, therefore,argued that as no loss was suffered by the owners after the outbreak ofwar, no damages should be recoverable after it commenced. The plaintiffs

82 I would, in addition to a claim for breach of contract for wrongful dismissal, also havea statutory claim for unfair dismissal: Employment Rights Act 1996.

83 [2007] 2 AC 353 (HL) [The Golden Victory].84 Indeed, on the Obligations Discussion Group I rashly said I thought it was wrong: see

<http://www.ucc.ie/law/odg/messages/070328b.htm> and the subsequent discussion (lastaccessed: 8 August 2008). Mea culpa.

85 The Golden Victory, above n 83 (Lord Scott, Lord Carswell and Lord Brown; LordBingham Cornhill and Lord Walker, dissenting).




sought to argue that damages should be assessed at the time of breach.Their argument was that the risk of a future war affected the market valueof the charter at that point, but as war was then far less certain, substantialdamages should be awarded representing the entire value of the charter attime of breach. By a bare majority, the House of Lords found for thecharterers, upholding the decision of the Court of Appeal.

The minority would have taken the time of breach as the time at whichdamages were crystallised based upon considerations of certainty andfinality. The quantum of damages would otherwise vary according to thechance of the moment at which judgment was given.86 Such an argument,if truly accepted, would lead to the conclusion that events occurring afterthe time of breach should never be taken into account in assessingdamages. This is clearly not the law. Certainty is a second order principleof justice. A rule which said that the damages for breach of contract arealways fixed at $100 would be even more certain, but nobody wouldaccept it as an acceptable rule.

The majority emphasised that damages are compensatory for actual losssuffered, and to ignore intervening events which have reduced this losswould lead to over-compensation. If, however, this is accepted, how are thecases where damages are assessed at the time of breach, ignoring subse-quent events, to be explained?

The answer is that The Golden Victory was a claim with respect toobligations to perform which had not accrued due at the time of repudia-tion. Exceptionally, the ability to claim damages can be accelerated. Once arepudiation has been accepted, damages for the loss of future performancecan be claimed at the time of acceptance, even though the time for suchperformance has not yet fallen due. However, what is accelerated is theability to claim damages, not the right to future performance itself. Thevalue of the right to performance must be quantified at the time perform-ance is due. In cases of the breach of an accrued contractual right, the rightis valued at the time of breach. Consequential losses are assessed at time ofjudgment. Where damages are awarded in lieu of a specific remedy, theright to performance is quantified at the time of trial. Where damages areclaimed with respect to future performance, the contractual right must bevalued at the time at which it would accrue to be performed. In TheGolden Victory, it was known that after the outbreak of the second GulfWar, the owner’s rights to further performance from the charterers wasvalueless because of the right to terminate. Therefore, regardless ofwhether the claim is seen as one for consequential loss or to the value ofthe charter, damages should have been and were discounted to take

86 See also GH Treitel, ‘Assessment of Damages for Wrongful Repudiation’ (2007) 123LQR 9.

196 Robert Stevens



account of the outbreak of war. The owners would never have had avaluable right to performance of the charter after the outbreak of war.

XI. CONCLUSION

In law, the purchase of my wife’s wedding ring has a happy conclusion.Substantial damages are payable. This is not because of my ‘reliance’ uponthe jeweller’s promise. This may be illustrated by some (fictional) figures.The contract price for the ring as promised was £500, but the market pricewas £2,000 (that is, I made a good deal, or at least I thought I did). Theplated ring delivered was only worth £50. My ‘reliance’ was no greaterthan the £500 I paid. The damages payable are £1,950, which is notequivalent to any sum I overpaid.

Similarly, the claim is not equivalent to a claim measured by the amountthat the seller made from my payment. This gain is not £1,950. The gainmade is £450, the difference between what was paid and what the ring isactually worth. A gain-based analysis does not explain the law either.

Whether the right to performance should be treated sufficiently seriouslythat its violation per se gives rise to a right to damages is a more difficultquestion to answer. Some rights are treated by the law as more important,and consequently more deserving of respect, than others. Some wrongs areactionable per se, some require the proof of consequential loss and someare not actionable at all. The breaking of promises supported by consid-eration falls into the first category, lies fall into the second, and insults fallinto the third. Some rights are more important than others, and conse-quently some wrongs are worse than others. However, there is no math-ematical formula or proof that enables us to demonstrate such ranking inthe abstract. In the end, we are each of us on our own in judging theimportance which ought to (morally) be attached to different rights. Itcould be the law that we treated the right to contractual performance inthe same way as our right not to be slandered, so that only consequentiallosses gave rise to a claim for damages. Indeed it could be the law thatcontractual rights and their infringement were not actionable in law at all,just like insults.87 The law is usually irrelevant in most contractualdisputes, as anyone who has got into a dispute with a builder will readilyunderstand. Gaming contracts were for many years void in England,88 butthis did not inhibit the enormous gaming industry. Betting companies veryrarely fail to pay out, unlike insurance companies who are in essentially thesame business but whose contracts are recognised by the law. Commerce isless dependent upon the law of contract than lawyers like to believe.

87 In Scotland, insults are actionable, as they were in Roman law.88 Gaming Act 1845, s 18; but see now Gaming Act 2005 ss 334–5.




At the margins, the law is there to provide determinate rules wheremorality is underdetermined. It may not be demonstrable whether thechoice the common law has made is correct or incorrect. Whatever themerits of different choices, there is too much law to the contrary for it tobe seriously argued that the law of damages for breach of contract is solelyconcerned with compensating losses (or stripping of gains). This has beenhidden from us for two reasons. First, in the majority of cases, but not all,the loss suffered is either greater than or the same as the value of the right.In travelling to the symposium at which this article was delivered mybaggage was lost by Air Canada. I had tempted fate by writing about thenon-delivery of goods. Damages calculated by reference to the marketvalue of my second-hand suits and shirts would clearly be inadequate. It isreasonable for me to incur the higher costs of replacement, and this will bethe usual sum awarded. Care must be taken that this normal position doesnot obscure the entitlement to damages even where such consequential lossis not incurred.

Second, misleading terminology has been adopted, with terms such as‘compensation’, ‘loss’ and ‘restitution’ being stretched to cover quitedifferent ideas with different incidental rules.

The first task of the academic lawyer is to explain the law so that itmakes coherent sense and to account for it in the best possible light. In thecommon law, we treat the right to contractual performance as sufficientlyimportant that when it is infringed, it gives rise to a claim for damagesrepresenting its value. My judgement, for what it is worth, is that this is(morally) appropriate. My wife agrees.

198 Robert Stevens



8

Estoppels and Rights-CreatingEvents: Beyond Wrongs and

Promises

ANDREW ROBERTSON*

THIS ARTICLE AIMS to explain the nature of the substantivedoctrines of estoppel as a source of rights and obligations. The mainfocus is on proprietary estoppel because that doctrine operates as a

substantive, independent source of rights in several jurisdictions includingEngland, Australia and Canada. The analysis extends to the broaderAustralian doctrine of equitable estoppel, which is essentially the same setof principles freed from the limitation that the assumption relied uponmust relate to an existing or expected interest in property. The expression‘equitable estoppel’ will be used to encompass both proprietary estoppeland the Australian doctrine, which is broader in scope than proprietaryestoppel but precisely the same in principle. The doctrines will be analysedby considering their place within a taxonomy of the law of obligationsorganised around rights-creating events. This is an exercise in interpreta-tive legal theory,1 rather than moral philosophy, and so its focus is onexplaining the pattern of liability in the case law rather than identifying amoral foundation for that liability.2

The doctrine of proprietary estoppel merits more attention than itreceives in theoretical literature on contract law.3 Proprietary estoppel is an

* I would like to thank Ben McFarlane and the participants in the ‘Exploring ContractLaw’ symposium for their helpful comments on earlier drafts of this article.

1 S Smith, Contract Theory (Oxford, Oxford University Press, 2004) 5–6; A Beever andC Rickett, ‘Interpretive Legal Theory and the Academic Lawyer’ (2005) 68 MLR 320, 321–5.

2 Whether the doctrines make ‘moral sense’ or whether a different pattern of liabilitymight make better moral sense are therefore outside the scope of this paper: see Smith, ibid, at243–4. A moral justification of the obligation created by a substantive doctrine of estoppelhas been advanced by M Spence, Protecting Reliance: The Emergent Doctrine of EquitableEstoppel (Oxford, Hart Publishing, 1999) ch 1.

3 See, eg, Smith, above n 1, at 233–4.



important source of rights arising from dealings that often involve prom-ises, and even exchanges of promises. The doctrine is regularly applied byEnglish and Australian courts in a wide range of contexts.4 It has beenaccepted that similar principles operate in Canada, although the doctrineseems to be applied far less frequently by Canadian courts.5 The Englishdoctrine of promissory estoppel, in contrast, receives an amount ofattention in the literature on contract law and contract theory that isperhaps undeserved, since the doctrine has a limited scope of applicationand is applied by the courts only occasionally.6 As Susan Bright and BenMcFarlane have observed, the notion that estoppel provides a cause ofaction only in property cases has no obvious justification and has persist-ently been criticised.7 A rationalisation is overdue.8 Until that occurs,however, the scholar interested in reliance-based liability arising fromnon-contractual exchange transactions, gratuitous promises and othercontexts in English law must turn to proprietary estoppel.

In England and Australia, proprietary estoppel is regularly applied bothinside and outside the familiar contexts of dealings between familymembers in relation to residential and commercial property, promises ofproperty made to caregivers and dealings between neighbours about rightsof way. The doctrine has also recently been applied, for example, in casesinvolving dealings between creditors concerning the priority of charges and

4 A recent study identified and discussed 18 English and 19 Australian cases over afive-year period in which remedies were granted to give effect to proprietary estoppel: ARobertson, ‘The Reliance Basis of Proprietary Estoppel Remedies’ [2008] The Conveyancerand Property Lawyer 295.

5 See, eg, Romfo v 1216393 Ontario Inc (2007) 285 DLR (4th) 512 (BCSC).6 A search of the ‘UK Cases Combined’ LexisNexis database (on 28 November 2007) for

cases in the preceding five years in which the words ‘promissory estoppel’ were used identifiedonly three cases in which the principle of promissory estoppel was applied: Msas GlobalLogistics Ltd v Power Packaging Inc [2003] EWCH 1393; Bottiglieri Di Navigazione SpA vCosco Qingdao Ocean Shipping Company (The ‘Bunga Saga Lima’) [2005] 2 Lloyd’s Rep 1(QB); Taylor v Rive Droite Music Ltd [2004] EWHC 1605 (Ch). In the last two of those casespromissory estoppel was an alternative justification for results reached on the basis of,respectively, collateral contract and interpretation of the contract terms. The principle wasalso applied in Business Environment Bow Lane Ltd v Deanwater Estates Ltd [2006] EWHC3363 (Ch), but that decision was overturned by the Court of Appeal: [2007] EWCA Civ 622.

7 S Bright and B McFarlane, ‘Personal Liability in Proprietary Estoppel’ [2005] TheConveyancer and Property Lawyer 14. As to the current position in English law, see BairdTextile Holdings Ltd v Marks & Spencer plc [2002] 1 All ER (Comm) 737 (CA).

8 See J Neyers, ‘A Coherent Law of Estoppel?’ (2003) 2 Journal of Obligations andRemedies 25, 31–3; D Nolan, ‘Following in their Footsteps: Equitable Estoppel in Australiaand the United States’ (2000) 11 King’s College Law Journal 202; B McFarlane and D Nolan,‘Remedying Reliance: The Future Development of Promissory and Proprietary Estoppel inEnglish Law’ (paper presented at the Obligations III conference, University of Queensland,July 2006) (cited with authors’ permission); B McFarlane, ‘The Problem of Pre-ContractualReliance: Three Ways to a Third Way’ (2006) Hauser Global Law Working Paper 12/06(http://www.law.nyu.edu/global/workingpapers/2006/ECM_DLV_015762) (last accessedNovember 2008).

200 Andrew Robertson



dealings between liquidators and company directors concerning the own-ership of property.9 The doctrine is also being applied to property otherthan land. In a recent English case proprietary estoppel was held to arisefrom dealings between partners in an accounting firm concerning beneficialinterests under a trust of a life insurance policy.10 In the well-publicisedlitigation relating to the copyright in the song A Whiter Shade of Pale itwas assumed that the claimant could be prevented from asserting hiscopyright interest in the song by way of proprietary estoppel (althoughultimately he was not).11 Moreover, Lord Hoffmann recently observed that‘There is no reason why the equitable rules of proprietary estoppel shouldnot apply to a patent in the same way as to any other property.’12 Thebroader Australian doctrine has been applied in a range of non-proprietarycontexts, although interestingly the great majority of the recent Australiancases involving the application of equitable estoppel have fallen within thescope of the orthodox doctrine of proprietary estoppel.13

I. WRONGS AND RIGHTS-CREATING EVENTS

Peter Birks has argued that the law of obligations can usefully beunderstood by reference to four categories of rights-creating events: ‘(1)contracts (consent), (2) torts (wrongs), (3) unjust enrichments, and (4)other events’.14 There are two significant and related problems with thecategory of wrongs. The first is that the category is too broad and abstract

9 Scottish & Newcastle plc v Lancashire Mortgage Corporation Ltd [2007] EWCA Civ684; Roufeil v Lusby [2003] NSWSC 1002; Hypec Electronics Pty Ltd v Mead (2003) 202ALR 688 (NSWSC), aff’d [2004] NSWCA 221.

10 Strover v Strover [2005] EWHC 860 (Ch).11 Fisher v Brooker [2008] EWCA 287.12 Yeda Research and Development Co Ltd v Rhône-Poulenc Rorer International

Holdings Inc [2008] 1 All ER 425 (HL) [22].13 A search of the Lexis-Nexis ‘Australian Commonwealth, State and Territory Caselaw’

database for cases over a five-year period from July 2002 to July 2007 identified only fivecases falling outside the scope of proprietary estoppel in which remedies were granted to giveeffect to equitable estoppel: EK Nominees Pty Ltd v Woolworths Ltd [2006] NSWSC 1172(involving dealings between a landlord and a prospective tenant of a supermarket); AnacondaNickel Ltd v Edensor Nominees Pty Ltd [2004] VSCA 167 (involving dealings between aproposed buyer and seller of shares); New Zealand Pelt Export Company Ltd v TradeIndemnity New Zealand Ltd [2004] VSCA 163 (involving dealings between a trader andtrade indemnity insurer concerning the application of the policy to sales not covered by itsterms); GEC Marconi Systems Pty Ltd v BHP Information Technology Pty Ltd (2003) 128FCR 1 (involving dealings between parties to a commercial contract involving a contractvariation); Gray v National Crime Authority [2003] NSWSC 111 (involving dealings betweenthe National Crime Authority and witnesses entering the witness protection program relatingto financial arrangements).

14 P Birks, ‘The Concept of a Civil Wrong’ in DG Owen (ed), Philosophical Foundationsof Tort Law (Oxford, Clarendon Press, 1995) 31, 33.

Estoppels and Rights-Creating Events 201



to have any explanatory power.15 For that reason, an obligation should beregarded as falling within the category of wrongs only if it cannot properlybe understood as falling within one of the other categories. The secondproblem is that while consent and unjust enrichment are events that giverise to primary rights, the category of wrongs is concerned, anomalously,with events that give rise to secondary rights. Birks accepted this anomalybecause, for the obligations that fall within this category, it is not possibleto identify at a generic level the events that give rise to the primary rights.16

Robert Stevens has offered a classification of the law of torts at the level ofprimary rights—the right to bodily safety, freedom of movement, reputa-tion, and so on—but suggests that the fact that many of these rights aregenerated by the event of birth demonstrates the limited utility of anevents-based categorisation.17

Although the Birksian events-based scheme has significant limitations,particularly when it comes to understanding the law of torts, it doesprovide a useful framework for analysing a substantive doctrine ofestoppel (that is, one that operates as an independent source of rights). Thequestions raised by the events-based taxonomy go to the heart of equitableestoppel, and a consideration of those questions provides significantinsights. Equitable estoppel can potentially be placed in three categorieswithin the events-based taxonomy. First, it could be classified within thelaw of wrongs. Whether equitable estoppel is properly classified within thelaw of wrongs depends on a negative answer to the question of whether itis possible to identify an event, or series of events, which gives rise toprimary rights that are recognised by the law prior to and independently ofany infringement of those rights.18 If it is possible to identify an event thatgives rise to primary rights, then estoppel does not belong within the law ofwrongs. It should be classified by reference to the event that gives rise tothe primary rights. Second, equitable estoppel could be classified, as Birkssuggested, as a consent-based obligation.19 This depends on whether thebasis of obligation under a substantive doctrine of estoppel is the makingof a promise or some other manifestation of consent. Third, equitableestoppel could be classified as an obligation that arises from an ‘other’rights-creating event. If equitable estoppel gives effect to primary rightsthat are created by an event other than consent, then equitable estoppelmust be regarded as a sui generis source of rights falling, in the Birksianscheme, within the ‘other’ category of rights-creating events.

15 Ibid, at 51.16 Ibid, at 46.17 R Stevens, Torts and Rights (Oxford, Oxford University Press, 2007) 287. Cf Birks,

above n 14, at 47.18 Birks, above n 14, at 50–51.19 P Birks, ‘Equity in the Modern Law: An Exercise in Taxonomy’ (1996) 26 University

of Western Australia Law Review 1.




A. Inconsistent Conduct as a Wrong

I have argued elsewhere, incorrectly, that equitable estoppel is best under-stood as part of the law of wrongs.20 That argument was based on the ideathat a wrong is committed when A, having induced B to adopt and actupon an assumption, acts inconsistently with the assumption withouttaking steps to ensure that B is not harmed as a result of that action.21

Many equitable estoppel cases are brought on the basis of a threateneddeparture from the assumption in question, rather than one that hasactually been committed. These cases can be reconciled with the view thatequitable estoppel deals with wrongs, on the basis that equitable estoppeloperates in some cases to restrain apprehended wrongs and in others toredress wrongs already committed. On this view, equitable estoppel has asignificant quia timet jurisdiction.

The crucial issue for an events-based taxonomy, however, is not whetherequitable estoppel responds to conduct that can be characterised as awrong. Rather, we must ask whether it is possible to characterise therelevant rights as arising from something other than the commission of awrong. The fundamental question is whether it is possible to identify anevent, or series of events, that creates primary rights recognised by the lawprior to and independently of any infringement of those rights. The notionthat equitable estoppel is concerned with ‘unconscionable conduct’ has ledto a view that no identifiable rights arise by way of estoppel until there hasbeen unconscionable conduct. In Ashton Mining Ltd v Commissioner ofTaxation, Merkel J held that ‘the rights created by an equitable estoppelcannot arise until there has been an unjust or unconscionable departure orthreat to depart from the assumption adopted and acted upon by the partyseeking to assert the estoppel’.22 This conclusion was based on two closelyrelated ideas that are commonly said to be foundational to estoppel. Thefirst is that the purpose of estoppel is to prevent an unjust departure froman assumption which has been acted upon. The second is that ‘the elementwhich both attracts the jurisdiction of a court of equity and shapes theremedy to be given is unconscionable conduct on the part of the personbound by the equity’.23 The purpose of estoppel and the equitablejurisdiction are not, on this view, enlivened unless there is a threat ofunconscionable conduct in the form of an unjust departure from the

20 A Robertson, ‘Situating Equitable Estoppel within the Law of Obligations’ (1997) 19Sydney Law Review 32, 57–60.

21 A Robertson, ‘Reliance, Conscience and the New Equitable Estoppel’ (2000) 24Melbourne University Law Review 218, 227.

22 [2000] ATC 4307 (FCA) [50] [Ashton Mining].23 Waltons Stores (Interstate) Ltd v Maher (1988) 164 CLR 387 (HCA) 419 (Brennan J)

quoted in Commonwealth v Verwayen (1990) 170 CLR 394 (HCA) 411–12 (Mason CJ) andSledmore v Dalby [1996] 72 P & CR 196 (CA) 208.




relevant assumption. Thus, in Commonwealth v Verwayen, Dawson J saidthat: ‘An estoppel will occur only where unconscionable conduct on thepart of one gives rise to an equity on the part of another.’24 This isconsistent with Kevin Gray and Susan Francis Gray’s view that the‘inchoate equity’ arising by way of proprietary estoppel ‘is brought intobeing when the landowner unconscionably sets up his rights adversely tothe legitimate demands of the estoppel claimant’.25

The issue in Ashton Mining was whether Ashton Mining Ltd (AM) wasentitled to a tax deduction with respect to a debt that AM had ‘writtenoff’. The debt had been owed to AM by its subsidiary, Ashton Gold Ltd(AG). A deduction was allowable under the Income Tax Assessment Act if‘all liability to pay any amount or amounts’ had been discharged.26

Whether liability to pay the debt had been discharged in this casedepended, inter alia, on whether AM ‘by its conduct, was estopped fromclaiming that the Ashton Gold debt had not been discharged’.27 Merkel Jdid not accept that AG had adopted or acted upon an assumption that thedebt had been discharged.28 In the end, however, he rested his decision onthe ‘additional difficulty with the estoppel case’,29 which was that ‘Therewas no departure, or threatened departure, by Ashton Mining from anyassumption adopted by Ashton Gold during the 1992 year of income.’30

The decision was therefore based on the notion that, without threatened oractual inconsistent conduct causing harm, no rights arise by way ofestoppel.

The question of principle that arises from this decision is whether, if Ahas induced B to adopt an assumption or expectation as to A’s futureconduct or B’s legal rights, which B has acted upon (in such a way that Bwould suffer detriment if A behaved inconsistently with the assumption),are the rights and obligations of the parties affected by that course ofevents if A has not departed or threatened to depart from the assumption?The issue becomes particularly clear if one takes the familiar familyproperty example, where A (the property owner) promises B (her son) thatA will allow B and his family to reside in a particular house owned by Aand that A will leave this property to B in her will. B and his family invest

24 Ibid, at 453–4.25 K Gray and SF Gray, Elements of Land Law, 4th edn (Oxford, Oxford University

Press, 2005) 967–8. Cf B McFarlane, ‘Proprietary Estoppel and Third Parties after the LandRegistration Act 2002’ [2003] CLJ 661.

26 Income Tax Assessment Act 1936 (Cth) ss 70B(2) and 159GP(1).27 Above n 22, at [43].28 It is interesting to note, however, that the following year the shares in AG were

purchased by a third party on the assumption that the debt had been discharged, so in slightlydifferent circumstances the estoppel issue could have been a live one, even without anyattempt by AM to depart from the assumption.

29 Above n 22, at [45].30 Ibid, at [54].




considerable personal effort and funds in renovating the house, reside in itfor 10 years upon the assumption that the house is theirs, and develop asignificant emotional attachment it. A has no intention of acting inconsist-ently with the assumption B has adopted. For the purpose of determiningA’s pension entitlements, however, should A be regarded as the solebeneficial owner of the property?

Once B has taken (potentially detrimental) action upon an assumptioninduced by A, there can be no doubt that A’s freedom of action isconstrained. The estoppel already governs the relationship between theparties because A is not, at this point, free to act inconsistently with theassumption that she has induced B to adopt and act on, at least withouttaking steps to ensure that departure from the assumption does not causeharm to B. As Ben McFarlane has argued, B must therefore be regarded ashaving rights against A by way of estoppel from the time reliance occurs.31

It follows that departure or threatened departure from the assumptioncannot be regarded as a prerequisite to an interest in land arising by way ofproprietary estoppel. On the facts of Ashton Mining, assuming somereliance by the debtor, it makes no sense to say that the creditor’s right torecover the debt remains unimpaired until the creditor attempts to exercisethat right, but the creditor will be prevented from recovering the debt if itattempts to do so. If the creditor will be prevented from recovering thedebt if it attempts to do so, then its right to the debt is clearly impairedbefore it so attempts.

This idea has now been recognised in several Australian cases.32 It hasbeen held in these cases that, for the purpose of determining pensionentitlements, a beneficial interest in land can effectively pass by way ofequitable estoppel in the absence of any attempt by the landowner to resilefrom the relevant assumption. In Repatriation Commission v Tsour-ounakis,33 for example, Mr and Mrs Tsourounakis owned a houseproperty which they allowed their son and his family to occupy. The sonwas induced to believe that he could occupy the property rent-free andconsider it his own and that the parents would leave the property to him intheir wills. The son and his family spent ‘a large amount of time, energyand money repairing and renovating the Property’ on the assumption thatit was theirs.34 Many years later, the question arose whether Mr and MrsTsourounakis should be regarded as the beneficial owners of the property

31 B McFarlane, ‘Blue Haven Enterprises Ltd v Tully & Another’ (2006) 1 Journal ofEquity 156, 159, building on McFarlane, above n 25, at 665–74.

32 In addition to the cases discussed in the text, see Kintominas v Secretary, Departmentof Social Security (1991) 103 ALR 82 (FCA); Sarkis v Deputy Commissioner of Taxation[2005] ATC 4205 (VCA); Secretary, Department of Family and Community Services v Wall[2006] FCA 863.

33 [2004] FCAFC 332.34 Ibid, at [50].




for the purpose of determining their entitlement to a veteran’s servicepension. The Repatriation Commission argued that since Mr and MrsTsourounakis had made no attempt to resile from the assumption, the sonhad ‘no cause of action that would result in his being recognised as havingany equitable claim or interest in respect to the Property.35 Spender, Kiefeland Emmett JJ held:

That contention is somewhat facile. The question is not whether Mr and MrsTsourounakis are threatening to act in an unconscionable manner. The questionis whether, if they did, Michael would be entitled in equity to restrain them fromdoing so. If he would, the value of the Property to Mr and Mrs Tsourounakismust be diminished to the extent that they would be required to compensateMichael as a term of avoiding any restraint by a court of equity. If a court ofequity would treat Michael as the beneficial owner of the Property, the value ofthe interest of Mr and Mrs Tsourounakis must be regarded as nil.36

The question whether equitable estoppel creates rights without inconsistentconduct is not only relevant for determining issues relating to taxation andsocial security entitlements. The question can have implications for therights and obligations asserted between individuals where the inchoateestoppel is relied upon by a third party. This situation arose in Young vLalic.37 Mrs Lalic induced her son (Mr Lalic) to believe that if he built ahouse on land she owned, she would allow him to reside in the houseindefinitely and would ultimately convey it to him. The son induced hisfiancée, Ms Young, to believe that the mother would permit Mr Lalic andMs Young to occupy the property together and would ultimately transfer itto them. On the faith of this assumption Ms Young paid $50,000 into thebank account of Mrs Lalic for the purpose of financing the construction ofthe house. The relationship between the son and Ms Young founderedbefore the house was complete. Although Mrs Lalic did not ever actinconsistently with the assumption she induced her son to adopt, MsYoung sought relief against her. Brereton J held that Ms Young could notassert an estoppel against Mrs Lalic because Mrs Lalic did not induce MsYoung to adopt the relevant assumption and did not know the money hadbeen paid into her bank account. But Brereton J held that Mr Lalic hadacquired an interest in his mother’s land by way of equitable estoppel, andthat interest could be conveyed, or pass by way of equitable estoppel, to athird party.38 Mr Lalic had, by his conduct, conferred ‘a subsidiaryequitable interest on Ms Young, which she can enforce directly against MrsLalic’.39

35 Ibid, at [51].36 Ibid.37 (2006) 197 FLR 27 (NSWSC).38 Ibid, at [83], citing Hamilton v Geraghty (1901) 18 WN (NSW) 152 (SC).39 Young v Lalic, above n 37, at [83].




Thus, although Mrs Lalic had not behaved inconsistently before MsYoung instituted proceedings,40 she had induced her son to adopt anassumption that the land would be conveyed to him. His reliance upon thatassumption gave him an interest in the land by way of estoppel. He in turnwas able to confer ‘a subsidiary equitable interest’ in the property on MsYoung by way of estoppel. Ms Young was able to rely on Mr Lalic’sestoppel against Mrs Lalic as a basis for establishing her own claim againstMrs Lalic. Although Brereton J did not discuss the fact that Mrs Lalic hadnot behaved inconsistently before the time the proceedings were instituted,the recognition that Mr Lalic had rights against his mother by way ofestoppel, and an interest in her land, is consistent with the decisionsdiscussed above, and is justified for the same reasons.

The Australian pension cases discussed above, along with Young v Lalic,show that a series of events consisting of assumption, inducement andaction in reliance gives rise to primary rights which are recognised by thelaw before the inducing party has engaged in any inconsistent conduct.These decisions must be right as a matter of principle. Once there has beendetrimental reliance by one party upon an assumption induced by anotherit is clear that the inducing party’s freedom of action is constrained. Theinducing party has an obligation to the relying party and the relying partyhas corresponding rights against the inducing party. The estoppel governsthe relationship between the parties before the inducing party engages inany inconsistent conduct. If inconsistent conduct causing harm is regardedas the wrongful conduct in proprietary estoppel, then it is clear thatproprietary estoppel is not part of the law of wrongs. That is becauseidentifiable, legally-recognised primary rights pre-date the commission ofany wrong.

B. Inducing a Change of Position as a Wrong

The analysis above depends on the notion that it is inconsistent conductthat constitutes the wrong. Proprietary estoppel could, however, beregarded as part of the law of wrongs if the inducement of a change ofposition, rather than inconsistent conduct, was regarded as the wrong.Peter Benson has suggested that ‘a reliance-based action for breach ofpromise should be recognized and approached as one variant of tort

40 In the proceedings brought by Ms Young, Mrs Lalic denied telling Mr Lalic that shewould transfer the land to him and ‘said that she did not propose to transfer land to MrLalic’: ibid, at [93].




liability for negligent statements or representations’.41 This follows WarrenSeavey’s view that the wrong in promissory estoppel is not in disappointingthe claimant’s expectation, but in causing his or her change of position:

Estoppel is basically a tort doctrine and the rationale of [section 90] is thatjustice requires the defendant to pay for the harm caused by foreseeable relianceupon the performance of his promise. The wrong is not primarily in deprivingthe plaintiff of the promised reward but in causing the plaintiff to changeposition to his detriment.42

This understanding, if correct, would square the idea that estoppel is basedon a wrong with the notion that rights arise by way of estoppel before anyinconsistent conduct. The crucial question, however, is whether causing theclaimant to change his or her position, in circumstances in which thatreliance is foreseeable, can be regarded as wrongful. The reason it cannot isthat there is nothing wrong, in itself, with inducing an assumption uponwhich you know another person will rely. Provided the assumption is madegood, the relying party suffers no harm as result of his or her reliance. Therelying party may even benefit in some situations.43 As Charles Goetz andRobert Scott have pointed out, a gratuitous promise gives the promisee avaluable piece of information about the future, which the promisee can useto his or her advantage by acting on the promise.44 If the promisee’sexpectations are fulfilled, the promisee stands to benefit from the advanceknowledge. Goetz and Scott called this ‘beneficial reliance’. This insightmakes it clear that inducing reliance cannot in itself be regarded aswrongful. It can only be considered wrongful if there is some reason todoubt the reliability of the assumption, and thus some prospect of harm tothe relying party.

If negligent conduct is to be understood as the basis of equitableestoppel, therefore, it would need to consist of a failure to warn theclaimant of the risk that the assumption might not be fulfilled. Thedefendant’s failure to warn the claimant can be seen as careless if areasonable person would have foreseen the circumstances which caused thedefendant to change his or her mind and would have taken steps to warnthe claimant about that risk. It is possible that some equitable estoppelcases can be understood this way. In Waltons Stores (Interstate) Ltd v

41 P Benson, ‘The Expectation and Reliance Interests in Contract Theory: A Reply toFuller and Perdue’ [2001] Issues in Legal Scholarship 5, 61. See also P Benson, ‘The Unity ofContract Law’ in P Benson (ed), The Theory of Contract Law: New Essays (Cambridge,Cambridge University Press, 2001) 118, 177.

42 WA Seavey, ‘Reliance upon Gratuitous Promises or Other Conduct’ (1951) 64 HarvardLaw Review 913, 926.

43 For an example, see A Robertson, ‘The Failure of Economic Analysis of PromissoryEstoppel’ (1999) 15 Journal of Contract Law 69.

44 C Goetz and R Scott, ‘Enforcing Promises: An Examination of the Basis of Contract’(1980) 89 Yale Law Journal 1261.




Maher,45 for example, one might be able to say that a reasonable person inthe position of the prospective tenant would have foreseen the risk that,because the tenant was reconsidering its retailing strategy, the leasetransaction may not go ahead, and would have taken steps to warn theprospective landlord of this risk before the landlord went ahead withbuilding work.

This, however, opens up factual questions that the courts do notconsider in the estoppel cases. Whether the defendant’s conduct can beregarded as careless must depend on what the defendant knew about thecircumstances that might (and ultimately did) lead the defendant todisappoint the claimant’s expectations, and what inquiries a reasonableperson would have made. In quite a number of cases, some of which arediscussed below, an equitable estoppel arises even though, at the time theassumption was induced and acted upon, the defendant was mistaken as tohis or her own legal rights. Basing estoppel on carelessness in inducing therelevant assumption would require the courts in those cases to considerwhether a reasonable person in the defendant’s position would haveinquired as to the true legal position. The courts in equitable estoppel casesdo not ordinarily inquire as to these issues so it is not possible to saywhether the defendants in estoppel cases behaved carelessly in inducing therelevant assumption. The notion of carelessness cannot therefore provide acompelling basis for understanding equitable estoppel. Accordingly, wecannot conclude that a defendant found liable on the basis of equitableestoppel committed a wrong in inducing the relevant assumption.

II. PROMISES AND RIGHTS-CREATING EVENTS

The analysis above shows that, while equitable estoppel cases often requirethe court to respond to a wrong, the wrong is not itself the rights-creatingevent.46 Following Dixon J in Grundt v Great Boulder Pty Gold MinesLtd, we may observe that a party who has relied upon an assumptioninduced by another has nothing to complain about ‘[s]o long as theassumption is adhered to.’47 We could equally observe, however, that apromisee under a contract has nothing to complain about provided thepromisor adheres to his or her contractual promise. In each case thestatement obscures the nature and source of the rights that are contra-vened. What is of interest to scholars of contractual obligations, andshould be of interest to scholars of reliance-based obligations, is not only

45 Above n 23 [Waltons].46 As noted in S Waddams, Dimensions of Private Law (Cambridge, Cambridge Univer-

sity Press, 2003) 79.47 (1937) 59 CLR 641 (HCA) 674.




the breach of duty itself but also the series of events that create theobligation that has been breached. We need, in other words, to focus onthe ‘primary superstructure above the wrong’.48 If equitable estoppel is notpart of the law of wrongs, the taxonomic question that arises is whether itis concerned with rights arising by way of a promise, consent or somesimilar event which would bring it within the same broad category ofobligation as contract.49 This is a useful question to discuss because itrequires us to consider important questions about the basis of the obliga-tion in estoppel.

An obligation arises by way of estoppel from a series of events,consisting of (i) the inducement by A of an assumption which is adopted byB, and (ii) such action on the faith of that assumption by B that B wouldsuffer detriment if A did not adhere to it. It is this chain of events(assumption, inducement, detrimental reliance) that creates rights in therelying party and a corresponding obligation in the person who inducedthe assumption. The making of a contract is a chain of events (offer andacceptance involving an exchange and manifesting an intent to be legallybound) which creates a legal entitlement to the benefits expected under thecontractual agreement. As a statement of legal principle, we can say thatthe foundational chain of events in estoppel does not give rise to anentitlement to the expectation, but does give rise to an entitlement toprotection from harm resulting from any inconsistent conduct by thedefendant.50 In contract, the failure to fulfil the promisee’s expectation is awrong. In estoppel, a wrong is committed when the defendant behavesinconsistently without taking steps to ensure that this does not cause harmto the relying party.51 The defendant commits a wrong by infringing theclaimant’s right not to be harmed by the defendant’s inconsistent conduct.

Having identified the rights-creating series of events in estoppel, we cannow consider whether equitable estoppel is best seen as essentially contrac-tual, or as a sui generis category of right and obligation which falls withinthe ‘other causative events’ category in the Birksian taxonomy of the law ofobligations.52 Stephen Smith has argued that ‘the best overall explanation’

48 Birks, above n 14, at 50.49 As to whether contract itself is best understood as a consent-based category of

obligation, see A Robertson, ‘On the Distinction between Contract and Tort’ in A Robertson(ed), The Law of Obligations: Connections and Boundaries (London, UCL Press, 2004) 87; ARobertson, ‘The Limits of Voluntariness in Contract’ (2005) 29 Melbourne University LawReview 179.

50 The principle is discussed at length in Robertson, above n 4. Peter Benson hasadvanced a justification for this difference between contract and estoppel: Benson, ‘The Unityof Contract Law,’ above n 41, especially at 175–6.

51 Cf McFarlane, above n 31, at 159, who argues that, in the proprietary estoppel cases,no wrong is committed when the defendant attempts to act inconsistently because theclaimant has already acquired ‘a right based on the need to protect [the claimant’s] reliance’.

52 See, eg, Birks, above n 19, at 8.




of the English doctrine of promissory estoppel is that it is a promissorydoctrine which can be seen as resting on the same foundations ascontractual obligations.53 Smith noted that this conclusion ‘relies in parton moral arguments’.54 A similar claim, though based more squarely onanalysis of the case law, was made by Edward Yorio and Steve Thel todemonstrate that promissory estoppel in the United States is concernedwith enforcing serious promises, rather than protecting reliance.55 JamesEdelman has also argued that, although equitable estoppel is ‘a creaturethat defies taxonomy’, it is a doctrine which is essentially concerned withthe enforcement of promises.56 Although Smith was principally concernedwith the English doctrine of promissory estoppel, the criteria he usedprovide a useful structure for assessing whether proprietary estoppel andthe broader Australian doctrine are essentially reliance-based or promise-based doctrines. The three criteria are: first, whether a promise is requiredto establish liability; second, whether detrimental reliance is required toestablish liability; and third, whether the remedies essentially involve theenforcement of promises.

A. Is a Promise Required?

A basic doctrinal difference between equitable estoppel and the principle ofpromissory estoppel described in section 90 of the Restatement (Second) ofthe Law of Contracts is that the former is based on an induced assumptionwhereas the latter requires a promise. It is unclear whether this is adifference of substance and whether it has any practical effect on therelative scope of the application of the doctrines.57 The essence of apromise, both in ordinary speech58 and in the Restatement,59 is a commit-ment. In terms of accepted doctrine, it is clear that a promise is not

53 Smith, above n 1, at 244. See also PS Atiyah and S Smith, Atiyah’s Introduction to theLaw of Contract, 6th edn (Oxford, Oxford University Press, 2005) 127.

54 Ibid.55 E Yorio and S Thel, ‘The Promissory Basis of Section 90’ (1991) 101 Yale Law Journal

111. For a response see R Hillman, ‘Questioning the “New Consensus” on PromissoryEstoppel: An Empirical and Theoretical Study’ (1998) 98 Columbia Law Review 580.

56 J Edelman, ‘Remedial Certainty or Remedial Discretion in Estoppel after Giumelli?’(1999) 15 Journal of Contract Law 179, 179. See also J Edelman, ‘Taking Promises Seriously’(2007) 45 Canadian Business Law Journal 399.

57 See Robertson, above n 20, at 44–5.58 The Oxford English Dictionary Online, <http://dictionary.oed.com> (last accessed: 2

November 2007) relevantly defines a promise as ‘A declaration or assurance made to anotherperson (usually with respect to the future), stating a commitment to give, do, or refrain fromdoing a specified thing or act, or guaranteeing that a specified thing will or will not happen’(emphasis added).

59 The Restatement (Second) of Contracts (1981) § 2 defines a promise as ‘a manifesta-tion of intention to act or refrain from acting in a specified way, so made as to justify apromisee in understanding that a commitment has been made’.




required to found an equitable estoppel. In Australia, the thresholdrequirement is generally framed in terms of the defendant inducing anassumption as to the defendant’s future conduct or the rights of theparties.60 Similarly, the English Court of Appeal has accepted as ‘the basicprinciples of proprietary estoppel’ that a person can be prevented fromasserting his or her legal rights if, short of a binding contract, the personmakes a promise that he or she will not insist on them, and:

Short of an actual promise, if he, by his words or conduct, so behaves as to leadanother to believe that he will not insist on his strict legal rights—knowing orintending that the other will act on that belief—and he does so act, that againwill raise an equity in favour of the other; and it is for a court of equity to say inwhat way the equity may be satisfied.61

Although the courts say that a promise is not required to found anequitable estoppel, one might nevertheless make the argument that apromise is required in practice. It might be argued, for example, that theonly assumptions that can reasonably be relied upon are those that arisefrom promises.62 Smith has questioned whether there is any conduct shortof a promise on which one can justifiably rely. He suggests that ‘a personwho refrains from making a promise is thereby indicating that she isreserving the right to change her mind’ and so a listener who relies on anon-promissory statement ‘does so at his own risk’.63 There is an obviousand well-recognised difficulty in distinguishing between conduct thatamounts to a promise and conduct falling short of a promise, since acommitment to behave in a particular way in the future is usually impliedrather than expressed. Nevertheless, one might question whether allconduct falling short of a promise necessarily communicates the reserva-tion of a right to change one’s mind.

Many proprietary estoppel cases arise from conduct that could properlybe characterised as a promise, in the sense that the defendant has expresslyor implicitly made a commitment to the claimant. Whether the defendant’sconduct can be characterised as a promise is an important question in anestoppel case because reliance on a clear promise is more likely to be

60 Waltons, above n 23, at 407, 428–9, 453, 458; Commonwealth v Verwayen, above n23, at 412–13, 444, 460–61, 487, 500.

61 Jennings v Rice (2002) [2003] 1 P & CR 8 (CA) [19] (Aldous LJ with whom RobertWalker LJ and Mantell LJ agreed), quoting Crabb v Arun DC [1976] Ch 179 (CA) 187 (LordDenning MR).

62 See S Smith, ‘The Reliance Interest in Contract Damages and the Morality of ContractLaw’ [2001] Issues in Legal Scholarship 1, 23. Cf Benson, ‘The Unity of Contract Law,’ aboven 41, at 174: ‘First, there must be a promise or some other representation by one party uponwhich the other may reasonably rely. An essential condition of such reasonableness is that thesecond party must reasonably be able to construe the first party’s intention as inviting relianceupon the promise or representation.’

63 Atiyah and Smith, above n 53, at 127.




regarded as reasonable. A promise may even be ‘perceived by reasonablepersons as inviting reliance’, as Benson has suggested.64 A promise is,however, neither necessary nor sufficient to found a proprietary estoppel.Although many proprietary estoppel cases are founded on conduct that canaccurately be characterised as a promise or commitment by the defendant,there is a second group of cases in which there is scope for argument aboutthis. These are the cases in which the language of assurance or commitmentis not expressly used, and it is arguable whether a commitment might havebeen implicit. There is, more significantly, a third group of cases in whichproprietary estoppels arise in circumstances in which it is not possible toimply any promise or commitment. Proprietary estoppel can arise from‘acquiescence by one party in the known expectation of the other partythat he has or will have a proprietary right or interest where the otherparty has acted to his detriment on that basis’.65 The cases applying thisprinciple clearly show that a promise is not required to establish aproprietary estoppel.66

We can begin with Stiles v Cowper, where a remainderman wasprevented from denying the validity of an invalid lease created by a lifetenant.67 The remainderman accepted rent from the tenant, and stood bywhile the tenant expended £5,000 rebuilding the house. Lord HardwickeLC said that while the acceptance of rent by itself would not bind theremainderman, ‘when the remainder-man lies by, and suffers the lessee …to rebuild, and does not by his answer deny that he has notice of it, all ofthese circumstances together will bind him from controverting the leaseafterwards’.68 As Oliver LJ later noted, it appears from the report that theremainderman did not know at the time he accepted the rent andacquiesced in the expenditure that the lease was invalid. He accepted therent because he ‘thought proper’ to do so.69 In those circumstances, it isnot possible to say that the landlord impliedly undertook to honour thelease. By accepting the rent and acquiescing in the expenditure, thelandlord simply confirmed the tenant’s assumption that the lease wasbinding. The landlord was bound by ‘all these circumstances’ not to actinconsistently with that assumption. Oliver LJ’s influential decision inTaylors Fashions Ltd v Liverpool Victoria Trustees Co Ltd confirmed thatan estoppel by acquiescence can arise even where the defendant was, at the

64 Benson, ‘The Expectation and Reliance Interests in Contract Theory,’ above n 41, 61.65 Scottish & Newcastle plc v Lancashire Mortgage Corporation Ltd, above n 9, at [44].66 As Waddams, above n 46, at 60–61 has observed, citing Ramsden v Dyson (1866) LR

1 HL 129; Spiro v Lintern [1973] 3 All ER 319 (CA); Crabb v Arun DC, above n 61;Waltons, above n 23.

67 (1748) 3 Atk 692, 26 ER 1198 (Ch) [Stiles].68 Ibid, at 1198.69 In Taylors Fashions Ltd v Liverpool Victoria Trustees Co Ltd [1982] QB 133 (ChD)

148.




time of the relevant conduct, mistaken as to his or her own legal rights.70

Where this occurs, the concept of promise has no explanatory power, asthe following recent cases show.71

In Scottish & Newcastle plc v Lancashire Mortgage Corporation Ltd,Lancashire Mortgage Corporation (LMC) advanced £30,000 to Pexmanon the security, inter alia, of a charge over Pexman’s house.72 Pexman wasat that time already indebted to Scottish & Newcastle (S&N) and, in fact,some of the money advanced by LMC went to reduce Pexman’s indebted-ness to S&N. S&N took a charge to secure the amount outstanding, andhad its charge registered before LMC’s. Pexman defaulted in repaymentand the proceeds of the sale of the house were insufficient to meet bothdebts. A proprietary estoppel arose which prevented S&N from assertingthe priority of its charge. The basis of the estoppel was that S&N knewthat LMC expected that its charge would have priority over that of S&N,and stood by and allowed LMC to advance money on the faith of thatexpectation.73 Mummery LJ expressly noted that this was ‘a case of passiveacquiescence by S&N rather than positive representation, encouragementor promise’.74

In Munt v Beasley a tenant improved a loft space on the mistakenassumption that the space was included in the lease.75 The landlord wasalso mistaken as to the subject matter of the lease, but was aware that thebuilding work was being undertaken and made no complaint about it. Itwas therefore held to be unconscionable for the landlord to assert his legalrights when he became aware of the true position. The Court of Appealsaid that ‘It would be unconscionable in this case, as Mr Beasley acqui-esced in the works and Mr Munt suffered detriment in executing them inthe belief that the loft was included in the Lease.’76 In a case such as thisthere is no basis for saying that the defendant made a commitment orpromise that his legal rights would not be enforced.77 Indeed, in cases suchas Munt and Stiles, where both parties are mistaken as to their legal rights,there cannot be an implied commitment or promise because the parties donot believe that there is anything to be making a commitment about.

The third recent English case is Strover v Strover.78 The partners in anaccounting firm effected ‘mirror policies’ of life insurance to protect one

70 Ibid.71 See also Roufeil v Lusby, above n 9, where company directors who allowed the

company to ‘order its affairs’ on the basis that it owned certain property were prevented fromdisputing that ownership.

72 Above n 9.73 Ibid, at [47].74 Ibid. Sedley LJ and Moore-Bick LJ agreed.75 [2006] EWCA Civ 370 [Munt].76 Ibid, at [44].77 For another recent example see Strover v Strover, above n 10.78 Above n 10.




another in the event that one of them should die during the life of thepartnership. The policies were intended to cover the expense that thesurviving partners would incur in having to purchase the partnershipinterest of the deceased from his estate. Each policy was held on trust, withthe two other partners nominated as beneficiaries. One of the partners diedafter he had retired from the partnership. At the time he retired he wronglyassumed that, following his retirement, the policy would enure for thebenefit of his wife and children. In fact the other partners remained thebeneficiaries under the trust. In reliance on that false assumption, thedeceased lost the opportunity to renegotiate the ‘mirror policies’ arrange-ment before he died. The surviving partners would have been likely toagree to correct the arrangement if the matter had been raised with thembecause they were similarly exposed with their own policies. Hart Jconcluded that a proprietary estoppel arose in the circumstances, but heldthat the remedy should reflect the fact that there was a 20 per cent chancethat the deceased would not have taken corrective steps had he not beenmistaken. Accordingly, the equity arising by way of proprietary estoppelwas held to be equal to 80 per cent of the proceeds of the policy. As inMunt and Stiles, since the defendants were mistaken as to their own legalrights, they cannot be said to have made any implicit promise or commit-ment.

It might be argued that, despite what the judges say, the basis ofobligation in the ‘acquiescence’ cases involving assumptions of rights isdifferent from that which underlies the ‘promise’ cases involving assump-tions as to the future conduct of the promisor. There are three problemswith this argument. The first is the difficulty of identifying an alternativebasis of liability in the acquiescence cases. The obvious candidate is unjustenrichment, particularly in light of the Privy Council’s confusion of theprinciples of unjust enrichment with those of proprietary estoppel in BlueHaven Enterprises Ltd v Tully.79 Birks once argued that the courts should‘distinguish in this context between the cause of action based on promisesand the cause of action based on free acceptance’.80 He said that ‘Wherethere is no promise or expectation-inducing conduct, only the cause ofaction in free acceptance should be recognised and only the restitutionarymeasure of recovery should be allowed.’81 In improvement cases such as

79 [2006] UKPC 17. A claim ‘based on unjust enrichment’ relating to the improvement ofland was rejected on the ground that the defendant did not acquiesce in the improvement ormake any relevant representation to the claimant. The case is discussed by McFarlane, aboven 31 and by K Low, ‘Unjust Enrichment and Proprietary Estoppel: Two Sides of the SameCoin?’ [2007] Lloyd’s Maritime and Commercial Law Quarterly 14.

80 P Birks, An Introduction to the Law of Restitution, rev edn (Oxford, Clarendon Press,1989) 293.

81 Ibid. See further N Hopkins, ‘Estoppel and Restitution: Drawing a Divide’ in E Cooke(ed), Modern Studies in Property Law (Oxford, Hart Publishing, 2003) vol 2, 145.




Stiles, it is possible to point to an enrichment which the defendant hasreceived at the expense of the claimant.82 There are, however, acquiescencecases such as Strover v Strover in which the defendants were not in anyrelevant sense enriched by the actions of the claimant. Moreover, where theacts of reliance do enrich the defendant, the remedies granted by way ofestoppel redress the detriment resulting from reliance, rather than restoringthe enrichment.83 In Munt, for example, the court had before it evidence asto the amount by which the tenant’s DIY loft-conversion increased thevalue of the defendant’s land,84 but ignored this in favour of a remedywhich was ‘proportionate to the expenditure of money and time on theconversion’.85 In Scottish & Newcastle plc v Lancashire Mortgage Corpo-ration Ltd, the defendant was clearly enriched, since some of the moneyadvanced by LMC (£20,000) was used to reduce the debtor’s liability toS&N.86 The effect of the estoppel, however, exceeded the amount of theenrichment (£20,000), since it gave LMC priority for the entire amountadvanced (£30,000). The estoppel operated to the extent of the claimant’sreliance interest, which happened in this case to coincide with its expecta-tion interest.

The second problem with attributing the cases involving assumptions ofrights to a different equity from those involving assumptions as to futureconduct is that in some fact situations it is difficult to distinguish betweenthem. In Waltons,87 for example, where negotiations for a lease wereconcluded and the documents had been signed by the landlord-claimantsand sent to the tenant-defendant for signature, the relevant assumptioncould be characterised in different ways. It could be characterised as anassumption that the tenant would sign the lease (an assumption as tofuture conduct), that the tenant had signed the lease (an assumption as toexisting fact) or that a binding agreement had come into effect between theparties (an assumption as to rights). It would be most unsatisfactory ifsignificant consequences flowed from the fine distinction between assump-tions as to rights and assumptions as to future conduct, in addition tothose which already flow from the fine distinction between assumptions offact (raising the evidentiary doctrine of common law estoppel by represen-tation) and assumptions as to rights or future conduct (raising thesubstantive doctrines of equitable estoppel).

82 As Low, above n 79, at 16, has noted, however, the courts in the acquiescence casestend to ignore the presence or absence of an enrichment.

83 Cf Waddams, above n 46, at 59, noting that enrichment is not always present and doesnot normally shape the remedy, but suggesting that the prevention of unjust enrichment is notirrelevant to proprietary estoppel.

84 Above n 75, at [5].85 Ibid, at [45].86 Above n 9.87 Above n 23.




The third problem with drawing a distinction between assumptions ofrights and assumptions as to future conduct is that assumptions as tofuture conduct can also arise by way of acquiescence. Waltons could beseen as an example of this. A more recent example is Hypec Electronics PtyLtd v Mead.88 Mr and Mrs Mead, the directors of Hypec, used thecompany’s money to purchase certain properties. It was not clear whetherthe money had been lent to the directors or the directors held theproperties on trust for Hypec. The Meads’ marriage broke down andHypec went into liquidation. The liquidator initially treated the propertytransactions as loans made to the directors by Hypec. Mr Mead sought touse the properties to fund expensive litigation disputing a debt for whichjudgment had been entered against Hypec in favour of a companyassociated with Mrs Mead. The liquidator allowed Mr Mead to incurconsiderable costs proceeding with the litigation on the assumption thatthe properties would be available to fund the litigation. By allowing MrMead to obtain orders for the sale of the properties in Family Courtproceedings between the directors, and by failing to reserve his right toassert Hypec’s interest in the properties, the liquidator was held to haveinduced the assumption that he would allow the properties to be used inthis way and would not assert the company’s interest. The liquidator wastherefore held to be estopped from asserting Hypec’s entitlement to theproperties.89 The relevant assumption was induced by acquiescence, but itclearly related to the liquidator’s future conduct rather than the director’sexisting rights in the properties.90 It would be extremely artificial to say inthese circumstances that the liquidator made a promise or commitment toact in a particular way. The estoppel arose because the liquidator had ledMr Mead to believe that he would behave in a particular way, namely thathe would allow the properties to be used to fund the litigation and notassert the company’s interest in them, and he knew that Mr Mead wasincurring substantial legal costs on the faith of that assumption.

Although the conduct on the part of the defendant that gives rise to aproprietary estoppel can in many cases accurately be characterised as acommitment or a promise, the doctrine cannot be characterised as apromissory doctrine because there is a significant minority of cases inwhich estoppels arise from conduct that clearly falls short of a promise.Nor can we conclude that the obligation in estoppel cases is voluntary orconsensual. Michael Pratt has suggested that the orthodox position is that

88 Above n 9.89 An order was made to ensure that the remedy was limited to Mr Mead’s reliance loss.

Relief was conditional on Mr Mead undertaking to inform the liquidator of the progress ofattempts to recover his costs from the other parties to the proceedings, and the liquidator wasgiven liberty to apply for an order for repayment of any amounts so recovered: ibid, at [199].

90 The assumption on which the estoppel was founded was that ‘the liquidator would notstand in the way of the orders being implemented’: above n 9 (CA), at [46].




‘an obligation is voluntary if and only if it depends for its validity on theintention of the obligor to acquire it, which intention is a positive reasonfor its existence’.91 There are three reasons why the obligation in propri-etary estoppel cannot be regarded as voluntary or consensual, even on thisrelatively undemanding definition.92 First, unlike contract, the principles ofestoppel do not require that the defendant manifest an intention to assumea legal obligation. Second, in cases like Munt, where the defendant believesthe claimant is entitled to do what he or she is doing, the defendant clearlycannot intend to assume an obligation. Third, even in those cases in whichthe defendant makes an express promise or commitment to behave in aparticular way in the future, the obligation that arises from estoppel is notcoextensive with the commitment or promise. Proprietary estoppel doesnot give rise to an obligation to do what the defendant has promised to dobut, as we will see shortly, simply gives rise to an obligation not to causeharm by behaving inconsistently.

B. Is Detrimental Reliance Required?

A crucial part of Yorio and Thel’s argument that the basis of promissoryestoppel under section 90 of the Restatement is promise, rather thanreliance, was that the elements of inducement and detriment are neithernecessary nor sufficient to establish liability.93 Similarly, Smith has ques-tioned whether detrimental reliance is a condition of liability in promissoryestoppel in England. There is good reason to doubt the role of reliance inthat doctrine, since detriment is not consistently required in promissoryestoppel cases.94 Smith’s scepticism about the detrimental reliance require-ment, however, seems to extend to the substantive doctrines of estoppel:

What is needed, in order to prove the reliance view [of promissory estoppel], isevidence of contract cases in which the courts denied estoppel arguments on thebasis of the reliance requirement. But it is not clear that any such cases exist.…In jurisdictions in which estoppel can be used as a cause of action, it should, intheory, be easier to find cases in which estoppel claims are denied because theplaintiff who should rely, did not …95

The example Smith seeks is of a situation in which a contractual obligationis waived by the plaintiff and not performed by the defendant, but in which‘it was proven that the defendant would not have performed regardless of

91 M Pratt, ‘Promises, Contracts and Voluntary Obligations’ (2007) 26 Law andPhilosophy 531, 535.

92 Cf Robertson, ‘The Limits of Voluntariness in Contract,’ above n 49.93 Yorio and Thel, above n 55, at 151–61.94 See E Cooke, The Modern Law of Estoppel (Oxford, Oxford University Press, 2000)

100–103; Neyers, above n 8, at 31–3.95 Smith, above n 1, 237–8 (footnote omitted).




the plaintiff’s waiver—in other words, that the plaintiff [this should be thedefendant] did not rely on the waiver’.96 In Brikom Investments Ltd vCarr97 it was argued unsuccessfully that the estoppel claim should failbecause one of the estoppel claimants would have acted as she did in theabsence of the representation. As Lord Denning MR explained, the courtsare unwilling to speculate about this counterfactual question where a partyenters into a transaction following the making of a promise or representa-tion by another:

It is no answer for the maker to say: ‘You would have gone on with thetransaction anyway.’ That must be mere speculation. No one can be sure whathe would, or would not, have done in a hypothetical state of affairs which nevertook place … Once it is shown that a representation was calculated to influencethe judgment of a reasonable man, the presumption is that he was so influ-enced.98

Reliance in this context means a sufficient causal connection between theassumption induced by the defendant and the detrimental change ofposition by the claimant. The law gives significant evidentiary advantagesto an estoppel claimant in relation to proof of this causal connection.Where the claimant has been induced to adopt an assumption andsubsequently acts to his or her detriment, a rebuttable presumption ofreliance arises.99 Moreover, reliance on the relevant assumption need notbe the claimant’s only reason for taking the detrimental action: it is enoughif it was a reason for acting.100 The same approach is used for causation inrelation to duress and misrepresentation.101 The point is not that reliance isunimportant but that whether a person has relied is a question relating tothat person’s state of mind, which cannot easily be proved or disproved. Ifdetrimental action has been taken following the inducement of an assump-tion then an attempt to identify the role played by each of the differentfactors that may have contributed to the decision to act is not likely to beilluminating. Lord Justice Cranworth put the point even more strongly inReynell v Sprye, where he said that

It is impossible so to analyze the operations of the human mind as to be able tosay how far any particular representation may have led to the formation of any

96 Ibid, at 237.97 [1979] QB 467 (CA).98 Ibid, at 482–3.99 Campbell v Griffin [2001] EWCA Civ 990, [25]; Wayling v Jones (1993) 69 P & CR

170 (CA); Greasley v Cooke [1980] 3 All ER 710 (CA) 713; Coombes v Smith [1986] 1 WLR808, 821.

100 Campbell v Griffin, ibid; Wayling v Jones, ibid, at 173.101 See, eg, A Robertson, ‘Partial Rescission, Causation and Benefit’ (2001) 17 Journal of

Contract Law 163, 168–73; R Bigwood, Exploitative Contracts (Oxford, Oxford UniversityPress, 2003) 347–51.




particular resolution, or the adoption of any particular line of conduct. No onecan do this with certainty, even as to himself, still less as to another.102

The decision of the English Court of Appeal in Campbell v Griffin showshow difficult it is to disentangle reliance from other motives.103 Theclaimant was a lodger who developed a close relationship with hislandlords. He undertook domestic work and provided personal care forthem, having been repeatedly assured that he had a home for life. Theclaimant said in his witness statement that he would have moved out had itnot been for the assurances, but admitted under cross-examination that hewas also influenced by his friendship with the landlords and a concern fortheir welfare. The trial judge found against the claimant on the basis thatthe detrimental action taken by the claimant had not been caused by hisreliance on the assurances made by the defendants. The decision wasoverturned by the Court of Appeal on the ground that the trial judge hadgiven insufficient weight to the presumption of reliance and had attachedtoo much importance to the claimant’s concessions about his mixedmotives. He had done much more than could be expected of even the mostfriendly lodger. The court held:

In cases of this sort it is inevitable that claimants should be asked hypotheticalquestions of the ‘what if’ variety but the court is not bound to attach greatimportance to the answers to such hypothetical questions. …it would do nocredit to the law if an honest witness who admitted that he had mixed motiveswere to fail in a claim which might have succeeded if supported by less candidevidence.104

It is clear from the above discussion that the courts will usually lack thefactual foundation necessary to scrutinise closely the issue of causation inestoppel cases. There can, however, be no doubt that detrimental reliance isa fundamental requirement of both proprietary estoppel105 and the broaderAustralian doctrine.106 A search of the LexisNexis databases confirms thatthe English and Australian courts routinely deny claims of proprietary (andequitable) estoppel on the basis that the claimant would not suffer materialdetriment as a result of his or her reliance if the defendant was allowed toact inconsistently with the relevant assumption. In many cases this is givenas a reason, amongst others, for denying liability.107 There are, however,

102 (1852) 1 De GM & G 660, 42 ER 710 (Ch) 728.103 Above n 99.104 Ibid, at [28]–[29].105 See, eg, Jennings v Rice, above n 61, at [21]: ‘There can be no doubt that reliance and

detriment are two of the requirements of proprietary estoppel’; Gillett v Holt [2001] Ch 210(CA) 232: ‘The overwhelming weight of authority shows that detriment is required.’

106 Waltons, above n 23, at 404, 429; Commonwealth v Verwayen, above n 23, at 413,429, 444, 455, 500. See Robertson, above n 20, at 44.

107 For example Gordan v Mitchell [2007] EWHC 1854 (Ch); Armitage Holdings Inc vDelahunty [2007] EWHC 1556 (Ch); Wakelam v Boardman [2007] NSWSC 567;




clear instances of cases in which the lack of detrimental reliance was theonly reason for rejecting the estoppel claim.108 In light of cases such asJennings v Rice109 and Sledmore v Dalby,110 it cannot be doubted thatdetrimental reliance is a crucial element of proprietary estoppel. Therequirement of detriment is also strictly applied in cases involving thebroader Australian doctrine of equitable estoppel. It is not difficult to findexamples of cases in which, even though a promise was made and actedupon by the claimant, the equitable estoppel claim failed because theclaimant’s action was held not to have been detrimental. A recent exampleis St George Soccer Football Association Inc v Soccer NSW Ltd.111 Theclaimants were soccer clubs excluded from the defendant’s ‘premier league’competition when it was reduced in size from 16 clubs to 10. Thedefendant had earlier told the claimants and other clubs that applicants forthe new competition were required to meet certain specified criteria.112 Thedefendant later changed its policy and did not adhere to the criteria inselecting the clubs that would participate. The court accepted that therepresentation was made and acted upon by the claimants in submittingtheir applications to join the new competition. The court continued:

But it cannot be said that they thereby acted to their detriment. The alternativecourse of action would have been to refrain from submitting applications. Hadthey done that, their ultimate position would have been the same as that they infact came to occupy, namely, non-participation in the new competition. Therequirements for the creation of an estoppel are therefore not established.113

There can be no doubt that detrimental reliance is an essential condition ofliability for both proprietary estoppel and the broader Australian doctrine.

Comptroller-General of Customs v Parker (2006) 200 FLR 44 (NSWSC); Yarmouth HarbourCommissioners v Harold Hayles Ltd [2004] EWHC 3375 (Ch); Bredel v Moore BusinessSystems Australia Ltd [2003] NSWCA 117; Hawker Pacific Pty Ltd v Helicopter Charter PtyLtd (1991) 22 NSWLR 298. See also Rosenberg v Fifteenth Eestin Nominees Pty Ltd [2007]VSC 101 and Mobil Oil Australia Ltd v Wellcome International Pty Ltd (1998) 81 FCR 475where the claims failed because any detriment suffered as a result of reliance by the claimantswas not proportional to the relief sought.

108 For example Hunt v Soady [2007] EWCA Civ 366; Fisher v Brooker, above n 11; StGeorge Soccer Football Association Inc v Soccer NSW Ltd [2005] NSWSC 1288; Etchison vANZ Executors and Trustee Company Ltd [2005] QSC 363; Sea Food International Pty Ltdv Theng Pew Lam, unreported, Federal Court of Australia, Cooper J, 27 February 1998,BC9800437.

109 Above n 61.110 Above n 23.111 Above n 108. Another example is Sea Food International Pty Ltd v Theng Pew Lam,

above n 108.112 Above n 108, at [141].113 Ibid, at [175].




C. Are the Remedies Promissory?

The third argument advanced in favour of the promise-based view ofestoppel is that the remedies typically involve the enforcement of thepromise.114 Smith asks:

Do these remedies protect the complainant’s ‘promissory’ interest in having apromise performed (as the promissory view asserts) or do they protect theclaimant’s ‘reliance’ interest in not suffering a reliance-based loss (as the relianceview asserts)? Here again, the evidence does not give a clear answer. In Englishlaw, which generally does not permit estoppel to be used as a cause of action, theissue of the appropriate measure of damages for an estoppel does not arise.115

Even in the ‘exceptional’ cases in which estoppel is permitted to operate asa cause of action, Smith suggests, the approach to relief supports thepromissory view of estoppel:

Under the doctrine of proprietary estoppel, for example, a landowner whorepresents to another that the owner’s land will be conveyed to him will berequired to do so if the other relies on that representation—even if the value ofthe reliance is less than the value of the land. The remedy in such cases is alwaysspecific performance—which, of course, is entirely consistent with promissoryconceptions of remedies.116

The recent English and Australian cases clearly contradict the promissoryconception of remedies.117 Both English and Australian law now requirethe relief granted in a proprietary estoppel case to be proportionate to thedetriment that has been or would be suffered by the claimant. In Jennings vRice, Aldous LJ described this as ‘the most essential requirement’ in thedetermination of relief.118 The Australian and English courts have givenexplicit consideration to that requirement in the great majority of therecent cases.119 Although the courts are careful to go no further than isnecessary to protect the claimant’s reliance interest, in most cases the onlyway to do so is by granting expectation relief, either in specie or inmonetary form. The principal reason for this is that the detriment that issuffered in proprietary estoppel cases tends to be of a type that is difficultto quantify. Common examples involve the performance of services thatthe claimant is not in the business of performing, emotional investment in a

114 In relation to s 90, see again Yorio and Thel, above n 55.115 Smith, above n 1, at 238.116 Smith, above n 1, at 238–9.117 This and the following paragraph draw on the analysis of the recent cases in

Robertson, above n 4. See further A Robertson, ‘Satisfying the Minimum Equity: EquitableEstoppel Remedies after Verwayen’ (1996) 20 Melbourne University Law Review 805; ARobertson, ‘Reliance and Expectation in Estoppel Remedies’ (1998) 18 Legal Studies 360.

118 Above n 61, at [36].119 See Robertson, above n 4.




family home developed through long-term residence, lost career opportu-nities and other life-changing decisions such as having children. Lostcommercial ventures and investment opportunities are also difficult toquantify.120 The courts cannot accurately compensate such detriment, butcan prevent it by fulfilling the claimant’s expectations. The difficulty thatthe courts experience in quantifying reliance loss in most proprietaryestoppel cases justifies approaching relief on the basis that the claimant hasa prima facie right to a remedy in the expectation measure. It also justifiesthe principle that the remedy should be proportional to the claimant’sreliance loss, rather than having a rule that the remedy should preciselycorrespond to the reliance loss.

In a significant minority of the recent proprietary estoppel cases both theEnglish and Australian courts have refused to grant relief in the expecta-tion measure on the basis that such relief would be disproportionate to theclaimant’s reliance loss.121 In some cases the courts have adopted amathematical approach, awarding the claimant compensation calculatedprecisely by reference to reliance loss.122 In Young v Lalic, for example, theclaimant contributed $50,000 to the construction of a house in theexpectation of a half-interest in property worth $800,000. Since shesuffered no other detriment, Brereton J held that fulfilment of her expecta-tions would be a disproportionate response, and her equity was held to besatisfied by a charge for her contribution, together with interest. In caseswhere the claimant’s reliance loss could not accurately be quantified, thecourts have made a ‘broad brush’ determination of an amount of compen-sation that is proportional to the detriment suffered by the claimant.123 Inone recent Australian case, proprietary estoppel had the effect of suspend-ing the defendants’ rights in order to give the claimant an opportunity toresume her original position. In Sullivan v Sullivan the claimant had givenup subsidised public accommodation for which she waited seven years onthe faith of a promise of a home for life.124 The effect of the estoppel wasto allow the claimant to remain in the promised house for a further sevenyears in order to give her an opportunity to return to public housing.

The routine awarding of expectation-based relief in estoppel cases cancomfortably be reconciled with the view that the doctrine is concerned toprotect against harm resulting from reliance rather than to fulfil promises.

120 For example Van Laethem v Brooker [2005] 2 FLR 495 (Ch).121 See, eg, Jennings v Rice, above n 61; Powell v Benney (2007) [2008] 1 P & CR DG12

(CA).122 Young v Lalic, above n 37; Repatriation Commission v Tsourounakis (2007) 158 FCR

214; Strover v Strover, above n 10. See also the reliance-based condition attached to theremedy granted in Hypec Electronics Pty Ltd v Mead, above n 9, mentioned in n 89 above.

123 Jennings v Rice, above n 61; Ottey v Grundy [2003] EWCA Civ 1176; Donis v Donis[2007] VSCA 89.

124 [2006] NSWCA 312.




In most cases there is no other way to ensure protection against harm,because the detriment cannot accurately be quantified. There is not,however, a similar promise-based explanation for the significant minorityof cases in which the courts grant more limited relief. We must thereforeconclude that the remedial goal is to protect against harm.

III. CONCLUSION

Although equitable estoppel routinely responds to conduct that could becharacterised as a wrong or apprehended wrong, equitable estoppel is notproperly seen as part of the law of wrongs. That is because in equitableestoppel, as in contract, it is possible to identify a series of events whichgive rise to primary rights that are recognised by the law prior to andindependently of any infringement of those rights. This significant featureof equitable estoppel distinguishes it from tort. In an events-based classifi-catory scheme, equitable estoppel is sui generis: the rights-creating event isdetrimental reliance on an assumption induced by another. To describeproprietary or equitable estoppel as ‘reliance-based,’ however, begs thequestion as to the basis of the obligation. As Stevens has observed, reliancealone is clearly an insufficient basis for liability.125 Although detrimentalaction in reliance is undoubtedly necessary for liability in equitableestoppel, and constitutes the final step in the rights-creating series ofevents, it is ultimately that which the claimant has relied upon that justifiesthe defendant’s liability. It is clear from the above discussion that it is notthe fact the defendant has made a promise that justifies his or her liability.A promise is neither necessary nor sufficient to establish liability. Nor canwe say that the defendant is responsible for having behaved carelessly ininducing reliance. The basis of equitable estoppel is an obligation not tocause harm through inconsistent conduct. Where A has played a role ininducing the adoption of an assumption by B, and A knows, intends, orshould reasonably expect that B may rely on the assumption,126 then thosecircumstances give rise to a duty on the part of A not to cause harm to B bybehaving inconsistently with the assumption.

125 Stevens, above n 17, at 15.126 See A Robertson, ‘Knowledge and Unconscionability in a Unified Estoppel’ (1998) 24

Monash University Law Review 115.




9

Lumley v Gye and the(Over?)Protection of Contracts

G H L FRIDMAN*

LUMLEY v GYE belongs in that group of nineteenth-century casesthat helped to form the fundamental principles of our moderncommon law.1 Notwithstanding the undeniable importance of the

case, the impression I have is that on the whole more attention has beenpaid, especially by academic writers, to the effect of the decision than tothe purpose behind it and the language in which it was expressed.2 Thatlanguage merits analysis, not least because the case is an illustration of‘judicial activism’. I mean by this phrase, which is so frequently employedin modern times particularly, but not exclusively, in relation to thejudgments of the Supreme Court of Canada, the fashioning by a court ofdecisions designed to achieve certain purposes thought desirable by thecourt, regardless of the correctness of those decisions in terms of thedoctrine of precedent.

Recently Lord Hoffmann declared that Lumley v Gye gave rise to whathe termed ‘accessory liability’, a new type of tortious responsibility.3 Ineffect, he was suggesting the liability recognised in the case emerged fromthe judges in the majority of the court much as the goddess Athene sprang

* I am greatly indebted to Stephen Pitel whose trenchant criticism of earlier draftspreserved me from many errors of taste, clarity and purpose. For those, if any, that remain,I am solely responsible.

1 (1853) 2 El & Bl 216, 118 ER 749 (QB).2 FB Sayre, ‘Inducing Breach of Contract’ (1923) 36 Harvard Law Review 663; CE

Carpenter, ‘Interference with Contract’ (1928) 41 Harvard Law Review 728; Note, ‘TortiousInterference with Contractual Relations in the Nineteenth Century: The Transformation ofProperty, Contract, Tort’ (1980) 93 Harvard Law Review 1510; D Dobbs, ‘TortiousInterference with Contractual Relationships’ (1980) 34 Arkansas Law Review 335; HPerlman, ‘Interference with Contract and Other Economic Expectancies: A Clash of Tort andContract Doctrine’ (1982) 49 University of Chicago Law Review 61; R Bagshaw, ‘Can theEconomic Torts be Unified?’ (1998) 18 Oxford Journal of Legal Studies 729; AP Simester andW Chan, ‘Inducing Breach of Contract: One Tort or Two’ (2004) 63 CLJ 132.

3 OBG Ltd v Allan (2007) [2008] 1 AC 1 (HL) [8]. Cf Lord Nicholls, ibid, at [172].



from the head of Zeus. As such it was a prime example, though of coursenot the only one in nineteenth-century England, of judicial activism.Whenever this occurs it is pertinent to inquire whether what the particularcourt did was justified in purely legal terms and, in so far as this isdeducible, what non-legal factors led the court to decide as it did.

I. THE DECISION ANALYSED

The issue raised by the demurrer pleaded by the defendant was whetherLumley was entitled to sue Gye for persuading Johanna Wagner toabandon her contract with Lumley to sing in his theatre so that she couldinstead sing for Gye in his. It is an indication of its novelty and difficultythat four months passed before four of the five judges of the Court ofQueen’s Bench (the Chief Justice, Lord Campbell, not participating)reconvened to give judgment. The majority in favour of allowing the actionto proceed consisted of Crompton, Erle and Wightman JJ. Coleridge Jdissented.

That what was involved in this case was not cut and dried had becomeclear when counsel addressed the court. In 1853 there unquestionablyexisted an action against someone who enticed a servant away from hismaster.4 Its origin was disputed. Crompton and Wightman JJ consideredthat it arose at common law. Coleridge J embraced the view that the caseswhich allowed the action derived from the provisions of the Statute ofLabourers of 1351, which supplemented the Ordinance of Labourers of1349.5 Both were passed because of the labour shortage resulting from theBlack Death.6 Coleridge J established, partly by reference to Year Bookcases, that the action came into being following the criminal liability setout in the statute.7 Even earlier, however, there was authority thatsupported liability where a master was deprived of his servant’s servicesbecause of a beating administered by the defendant. The basis of thisliability was the violence employed.8 Only after 1351 did there emergeliability for such interference without the use of violence, merely byseducing the servant away from the original employer.

The proposition that at common law an action lay at the suit of a masteragainst one who caused the breach of the master’s contract with hisservant, either by harming or killing the servant or by enticing him away,seems to conflict with the fundamental doctrine of privity of contract. InBowen v Hall Brett LJ clearly indicated that, as regards the origins of the

4 Sayre, above n 2, at 663–72.5 25 Edw III and 23 Edw III respectively.6 Dobbs, above n 2, at 340 note 27.7 Cf Sayre, above n 2, at 663–72.8 Dobbs, above n 2, at 338–40.

226 G H L Fridman



action for enticing a servant away from his master, the view of Coleridge Jin Lumley v Gye, explained above, was preferable to that of the majority.9

Those narrow origins did not support the broader decision in Lumley vGye. The true ratio decidendi of that case, the basis on which it could besupported and endorsed, was said in Bowen v Hall, much more broadly, tobe that:

wherever a man does an act which in law and in fact is a wrongful act, and suchan act as may, as a natural and probable consequence of it, produce injury toanother, and which in the particular case does produce such an injury, an actionon the case will lie.10

This was stated to have been derived from the much earlier case of Ashbyv White.11 However that decision was not cited, nor is this ratio to befound, in the judgments of Crompton and Erle JJ in Lumley v Gye. Aversion of this ratio, but without attribution to Ashby v White, doesappear in the judgment of Wightman J as one, but not the only nor thegoverning, reason for permitting the action. Wightman J seems to concurwith, and rely on, the ratio in the judgment of Crompton J, namely that, bywhat Crompton J called ‘strict analogy’ with the master–servant situationreferred to earlier, the same action could be brought where the contract inissue was not a contract of service.12

Erle J gave an entirely different reason for allowing Lumley’s action toproceed. He responded to the objection that, since actions for breach of acontract of hiring were based on no principle, such actions could not beextended beyond existing precedents, which related to contracts respectingtrade, manufactures or household service but not performance at a theatre.His response was that actions for procuring the breach of a contract ofhiring were indeed based on a principle. That principle, he said, was thatthe injured party had a right to the service and ‘the procurement of theviolation of the right is a cause of action’.13 From this he proceeded to themore general notion that the violation of a right was a cause of actionwhere the violation was an actionable wrong, as in a violation of a right toproperty or personal security. From this he concluded that ‘when thisprinciple is applied to a violation of a right arising upon a contract ofhiring, the nature of the service contracted for is immaterial’.14 Therefore itdid not matter that Joanna Wagner was not a servant in the strict legalsense.

9 (1881) 6 QBD 333, 339–40.10 Ibid, at 337.11 (1703) 1 Holt KB 524, 90 ER 1188.12 Lumley v Gye, above n 1, at 229.13 Ibid, at 232. Cf Mogul SS Co Ltd v McGregor Gow & Sons (1889) 23 QBD 598, 614

(Bowen LJ).14 Ibid.

(Over?)Protection of Contracts 227



Although the wrongfulness of the act was stressed in Lumley v Gye asbeing the basis of liability, ‘malicious behaviour’, the nature of which wasundefined and unexplained, was also mentioned as possibly relevant. Yetthe court did not countenance the view that ‘malice’ made actionable thatwhich otherwise was not actionable. Such an effect of malice was negated,in Bowen v Hall,15 by Coleridge J’s son, John Duke Lord Coleridge.16 Inthat case the majority upheld and approved Lumley v Gye, over the dissentof Lord Coleridge who, like his father before him, disagreed with thedecision. In arriving at this conclusion Brett LJ relied on the wrongfulnessin law and fact of what was done rather than on malicious behaviour.17

However, Brett LJ, by then Lord Esher MR, brought malice back into thepicture when, in his dissenting judgment in Mogul SS Co Ltd v McGregorGow & Sons, he suggested that liability under Lumley v Gye involvedacting with the ‘malicious intention’ of interfering with the plaintiff’sright.18 By malice he meant ‘a malicious motive’. Later, in Temperton vRussell it was clearly stated that the Lumley v Gye tort could be committedeven though the relevant contract was not one under which personalservice was provided by one party to the other.19 According to Lord Esher,inducing or procuring the breach of a contract was wrongful because itwas done ‘maliciously’.20 The other members of the court, AL Smith andLopes LJJ, did not use the same language but the general tenor of theirjudgments was to the same effect.21

The net result of these three cases was to create a new tort that girdedcontracts with a protective shield similar to, but more real than, the magicfire surrounding Brunnhilde. What the courts did was to impose liabilityfor acts that were held to be impermissible, in the words of one Americancommentator, ‘only because the defendant’s purpose is thought insuffi-ciently laudable’.22 Why was this? One reason sometimes offered, andperhaps more often just assumed, says the same commentator, ‘is thatinterference with a contract should produce liability because it is wrong tointerfere. This is, however, very much the same as saying that it is wrongbecause it is wrong.’23 In other words, the conclusion reached by the courtswas foregone. It was dictated by the very nature of the question, given thecontext. My own view is that these criticisms have considerable force.

15 Above n 9, at 342–3.16 The first Lord Chief Justice of England: W Holdsworth, History of English Law

(London, Methuen & Co Ltd, 1965) vol 15, 460–66.17 Bowen v Hall, above n 9, at 338.18 Mogul SS Co Ltd v McGregor Gow & Sons, above n 13, at 608–609.19 [1893] 1 QB 715.20 Ibid, at 728.21 Ibid, at 730, 732–3.22 Dobbs, above n 2, at 343.23 Ibid.

228 G H L Fridman



What may have contributed to this idea that interference with a contractwas wrong in itself is the frequently repeated notion of ‘the sanctity ofcontract’. Such language seems to elevate the mundane, though useful,concept of contract into something more sublime. As a recent Englishcommentator remarked, ‘the phrase is all very well, but contract law islargely about commerce, not holiness’.24 He underlined the point furtherby saying: ‘Contract law is a public service the state offers to people whowant to use it—rather like the National Health Service.’25

What Lumley v Gye did, said this author, was to turn into a tort theeveryday commercial act of offering someone a better deal, enshrining arule that was inimical to competition.26 To which it might be added thatcompetition could be considered an integral part of a free society, so thatanything thought to hinder competition, such as the doctrine emanatingfrom Lumley v Gye, might be undesirable. Hence, perhaps, the suggestionby Sayre in 1923 that ‘It will not do to make a fetish of this tort remedy forthe better protection of contractual rights.’27

Lumley v Gye made a profound change in the law. It gave birth to thetort of procuring or inducing a breach of contract. This, in turn, spawnedthe tort of wrongful interference with trade or business by the use ofunlawful means. Beyond these developments in the law of torts, in creatingaccessory liability for breach of contract the decision also had conse-quences for the concepts of contract, contractual rights and privity. Thesubsequent development of the Lumley v Gye tort, I suggest, brings to thefore the clash between morality and legal rights.

Before delving into these matters, however, one intriguing aspect of thecase merits mention. The action was permitted to proceed as a result of thedecision of the Court of Queen’s Bench. At the trial before a special jury,the plaintiff Lumley lost because, as Waddams has explained, he could notestablish the malicious conduct required on the part of the defendant forthere to be liability.28 This, in turn, was because the defendant believed, orat any rate convinced the jury that he believed, that the contract betweenthe plaintiff and Wagner had ceased to exist or had been terminated legallyby Wagner because the plaintiff had not paid her an agreed advance. Thistrial was presided over by Lord Campbell, the Chief Justice of the Court ofQueen’s Bench. As previously noted, he did not participate in the earlierproceedings arising from the demurrer. One might consider his absence

24 D Howarth, ‘Against Lumley v Gye’ (2005) 68 MLR 195, 202.25 Ibid, at 203.26 Ibid, at 202.27 Sayre, above n 2, at 686.28 S Waddams, Dimensions of Private Law (Cambridge, Cambridge University Press,

2003) 36–8.




from the court on that occasion, given the importance and novelty of theissue at stake, to be somewhat odd. Speculation as to his reasons, however,is idle.

If the role of Lord Campbell in this litigation is something of a mystery,what was done by Crompton, Erle and Wightman JJ is extremely clear.They chose to enlarge the scope of earlier decisions, to ignore the historicalrelevance of violence or criminal activity contrary to the 1351 statute asforming the basis of liability, to consider that they were not confined by thelanguage of that statute, and to introduce the idea of malicious behaviourand the concept of violation of rights as grounds for creating liability.Their decision was revolutionary. There was no pragmatic reason orjustification for their decision. The Wagner situation did not give rise to thepossibility that she would not be able to compensate Lumley—one reasongiven by Crompton J for allowing the action, with which can be contrastedthe suggestion in a difficult and confusing passage in the judgment of ErleJ that the only recourse for the innocent victim of a contract procured to bebroken is to sue the contract-breaker, not the procurer of the breach.29

Lumley could have sued Wagner for breach of contract and been compen-sated by damages.

In fact, as the court also knew, Lumley had successfully sought aninjunction against Wagner to prevent her from appearing for Gye, theeffect of which would be to force her to appear for Lumley or not appearat all. Why then did he also wish to sue Gye? Could he have obtaineddifferent damages from those recoverable from Wagner, as Crompton Jsuggested?30 Was the real reason for the action the vendetta or rivalrybetween these two theatre managers? If so, why did the Court of Queen’sBench consider its judicial duty was to create a remedy that would enableone rival to sue the other when there was no contractual nexus betweenthem? Was Lumley’s action motivated by spite or ill will? Was this whysome members of the court introduced the idea of malicious behaviour asan element of liability in this case? In effect, was the court making a rule togovern the precise situation in Lumley v Gye, couching it in language thatwas of more general application? Was this, therefore, an example of hardcases making bad law?

Reading between the lines, and despite the talk in the judgments aboutan analogy between the situation in this case and the master-servantrelationship leading to legal consequences for enticement, it would seemthat the court was enunciating a policy to be adopted and enforced by thelaw: to protect contracts like that between Lumley and Wagner and ensurethat they are upheld and performed. As we now know, in later decisions

29 Lumley v Gye, above n 1, at 232.30 Ibid, at 230.

230 G H L Fridman



these aims were extended to cover not only contracts of service or forservices but all contracts, of whatever nature. As we further now know, theprotection provided by the law extends (a) to cover acts that prevent theperformance of a contract even though the defendant has not induced orprocured one party to commit a breach, as long as he or she has broughtabout a breach31 and (b) to include within its scope acts which indirectlybring about the inducing or procurement of a breach.32 Such extensions ofthe original judgment are questionable. The first may be unnecessary, sincethe conduct involved might be actionable on other tortious grounds, suchas trespass. The second seems to come close to creating liability forconduct with only a remote connection to the ultimate breach of contract.Indeed the application of this extension of Lumley v Gye has causedproblems.

It may be surmised that Crompton, Erle and Wightman JJ little knewwhat a can of worms they had opened when they held that Lumley had agood cause of action against Gye. They could not foresee the confusion,recently demonstrated by Lords Hoffmann and Nicholls,33 between theLumley v Gye tort and the condemnation of interference not directlyconcerned with causing the breach or non-performance of a contract butrather aimed at disrupting the economic life of the plaintiff. Where theLumley v Gye tort, in its original form, did not entail the use of unlawfulmeans or acts, the newer tort made such means or acts an integral part ofliability. The initial irrelevance to liability under Lumley v Gye of anunlawful act to bring about the breach of contract is perhaps the mostremarkable aspect of the case. It seems to indicate, even underline, thetremendous legal significance of contracts. To cause a breach of contract,despite the absence of any otherwise criminal or tortious act, was sufficientfor liability. Contracts were to be taken seriously.34

This is not to say that prior to 1853 contracts were regarded lightly—quite the contrary. The nexus between two or more parties created by acontract gave rise to an important personal relationship. Indeed thepersonal nature of contract was a shibboleth of the law. The doctrine ofprivity came to maturity in the early nineteenth century. According to thisdoctrine, contracts concerned the parties thereto and no one else. An

31 GWK Ltd v Dunlop Rubber Co Ltd (1926) 42 TLR 376 (KB).32 DC Thomson & Co Ltd v Deakin [1952] Ch 646 (CA).33 In OBG Ltd v Allan, above n 3.34 Another important aspect of Lumley v Gye might be the emergence of the notion that

the intentional causing of economic, as contrasted with physical, harm or damage could per sebe the basis for liability in the absence of a legitimate reason or excuse. This is intimatelyconnected with the idea that certain types or instances of commercial competition were notconsidered valid behaviour and so were not to be countenanced. A similar attitude evolvedwith respect to economic duress making a contract voidable, if not void. In this context, too,the courts were faced with the need to differentiate valid from invalid behaviour, namelypressure that affected the making of an agreement.




important exception, operating long before Lumley v Gye, revolvedaround the relationship of master and servant. One who prevented aservant from fulfilling his obligations to his master by wrongfully inflictingan injury on the servant could be made liable to the master for his resultantloss.35 This did not apply except where it was a servant who was injured. Itdid not extend to contracts for the provision of services as distinct fromcontracts of service.

The reason for such liability, according to Kitto J of the High Court ofAustralia in Attorney-General (NSW) v Perpetual Trustee Co,36 was thatsuch conduct constituted a wrongful invasion of the quasi-proprietary righta master was considered to possess in respect of the services the servantwas under an obligation to render.37 Several arguments supporting theview of Kitto J were subsequently adumbrated by the High Court ofAustralia in Zhu v Treasurer of New South Wales,38 following thejudgment of Jordan CJ of New South Wales in Independent Oil IndustriesLtd v Shell Co of Australia Ltd,39 which had analysed the nature of thedefence of ‘justification’ in relation to inducing a breach of contract interms of a superior ‘property’ right.40 That approach was approved andendorsed in Zhu. It is an approach that severely limits the scope of thedefence of justification, confining it to inducing a breach of, or interferingwith, a contract when the alleged wrongdoer was merely protecting his orher own superior right of property, as explained by the English Court ofAppeal in Edwin Hill & Partners v First National Finance Corp plc.41

II. THE DEFENCE OF JUSTIFICATION

In reaching its conclusion in Zhu the High Court of Australia rejected thesuggestion of a wider scope to justification involving protection of contrac-tual rights equal or superior to the rights of the plaintiff. The court, likeJordan CJ earlier, following Lord Halsbury LC in South Wales Miners’Federation v Glamorgan Coal Co Ltd,42 also seems to have rejected theidea that justification could be founded on moral or religious grounds oron any duty or moral right to tender advice.

In these cases the courts seem to be declaring incorrect the decision ofRussell J in Brimelow v Casson, in which the defendants were not liable

35 G Jones, ‘Per Quod Servitium Amisit’ (1958) 74 LQR 39. See also the discussionaccompanying n 8.

36 (1952) 85 CLR 237 (HCA) 294–5.37 Cf Lumley v Gye, above n 1, at 232 (Erle J).38 (2004) 218 CLR 530 (HCA) 573–7 [Zhu].39 (1937) 37 SR (NSW) 394.40 Cf Bagshaw, above n 2, at 735.41 [1989] 1 WLR 225 (CA).42 [1905] AC 239 (HL) 244–5.

232 G H L Fridman



because they had acted out of a sense of morality in order to preventactresses employed by the plaintiff from having to resort to prostitution tomake up for their meagre salaries.43 If Lord Halsbury, Jordan CJ and theHigh Court of Australia are correct, it follows that, as far as liability underLumley v Gye is concerned, morality is irrelevant. It, in Koko’s words, ‘hasnothing to do with the case’.44 Only legal, proprietary or perhaps quasi-proprietary rights are pertinent.

If this is indeed the law I find it appalling. Even John Austin, I venture tosuggest, would be upset. In this regard I should refer to a recentcommentator who wrote that: ‘The case against Lumley is that it assumesthat there is something inherently wrong with persuading someone tobreach a contract, whereas such persuasion can often be morally justified,as well as being economically desirable.’45 This commentator would likethe law to take a broader, more liberal approach to justification, so as toconfine the operation of Lumley v Gye. He advocates a ‘maximalistapproach’ which would require consideration of whether it was ‘fair, justand reasonable’ for Lumley v Gye to apply, which would tend to make theoperation of the tort somewhat akin to the operation of the tort ofnegligence. Were this approach to be accepted, the courts would have todetermine whether (a) in order to succeed in a claim, the plaintiff wouldhave to show that it was fair, just and reasonable to allow the action, or (b)to avoid liability, the defendant would have to prove that to allow theaction would not be fair, just and reasonable. In other words, is justifica-tion a defence or is its absence an essential ingredient of prima facieliability?

Regardless of the ultimate resolution of this question, the meaning ofjustification as a response to an allegation of inducing a breach of contract,in light of the explanation referred to above and the moral approachinvoked in Brimelow v Casson, requires that reference be made to thelanguage of Romer LJ in Glamorgan Coal Co Ltd v South Wales Miners’Federation.46 He said that whether interference was justified depended on:

the nature of the contract broken; the position of the parties to the contract; thegrounds for the breach; the means employed to procure the breach; the relationof the person procuring the breach to the person who breaks the contract;and . . . the object of the person in procuring the breach.47

This language, which might be interpreted as opening the way to justifyinginducement of a breach of contract on moral grounds, was cited and

43 [1924] 1 Ch 302.44 WS Gilbert and A Sullivan, ‘The Mikado’ in The Annotated Gilbert and Sullivan—Vol

1 (London, Penguin, 1982) Act II, line 646.45 Howarth, above n 24, at 224.46 [1903] 2 KB 545 (CA), aff’d [1905] AC 239 (HL).47 Ibid, at 574.




approved, but interpreted differently, by Gale CJ in Posluns v TorontoStock Exchange.48 In his view there had to be a legal basis for intervention,such as a statute or contract, or some significant social reason. A moralbasis for intervention would not be enough unless joined with thepromotion of a legitimate trade union or other interest that advanced asocial policy. The language of Gale CJ, I suggest, might save Brimelow vCasson from being overruled. But the language of Jordan CJ, or thejudgment in Zhu, and the language of English judges Darling J and BuckleyLJ, cited and discussed in Zhu, will not.

The search for the meaning of justification in regard to the Lumley vGye tort, the effect of which would be to limit materially the scope of suchliability and to control the extent to which the law protects otherwise validcontracts, may arguably be said to still be ongoing. In this respect, I think,enough has been said to indicate that the modern development of Lumley vGye raises serious questions as to the role of morals and morality in thedetermination of legal doctrine.

This is not the place to discuss that issue. My purpose in pursuing thisexamination of the background to, and effects of, the decision in Lumley vGye was to consider the case from the point of view of judicial activism. Ican only hope that I have managed to reveal that what was done in 1853was something that was contrary to precedent and a step in the wrongdirection, and, in the end, was capable of producing undesirable conse-quences. The point was made, at the time, by Coleridge J, who should havethe last word:

It seems to me wiser to ascertain the powers of the instrument with which youwork, and employ it only on subjects to which they are equal and suited; andthat, if you go beyond this, you strain and weaken it, and attain but imperfectand unsatisfactory, often only unjust, results. But, whether this be so or not, weare limited by the principles and analogies which we find laid down for us, andare to declare, not to make, the rule of law.49

III. ADDENDUM

Although not strictly necessary to the preceding argument, some informa-tion about the personalities involved as counsel or judges in Lumley v Gyeis not without interest.

About counsel for the plaintiff, Cowling, I know nothing. He does notappear to have been one of the leaders of the Bar, nor did he achieveappointment to one of the common law courts. Very different was thecareer of counsel for the defendant, James Willes. He was described by Sir

48 (1964) 46 DLR (2d) 210 (Ont HC) 270–73, aff’d (1965) 53 DLR (2d) 193 (Ont CA),aff’d [1968] 1 SCR 330, (1968) 67 DLR (2d) 165.

49 Lumley v Gye, above n 1, at 269.

234 G H L Fridman



William Holdsworth as ‘perhaps the most learned common lawyer of hisday’.50 Not long after he appeared for the defendant in Lumley v Gye hewas appointed a Justice of the Court of Common Pleas. As a judge he wasresponsible for some highly important decisions, in which he revealed apropensity for innovation that is to be contrasted with his affirmation ofthe status quo in his argument in Lumley v Gye.51 Undoubtedly he wouldhave attained higher judicial office. Unfortunately he suffered from heartdisease, gout and insomnia, and in 1872, following a nervous breakdownafter a heavy assize at Liverpool, he shot himself.

Justice Crompton, after practising in the Court of Exchequer, became ajudge of the Court of Queen’s Bench in 1852. There he proved himself tobe, according to Holdsworth, ‘a very sound lawyer and a good judge’.52

Justice Erle moved after two years in the Court of Common Pleas to theCourt of Queen’s Bench in 1847, but returned to the former court in 1859as Chief Justice. Several years later he resigned and became a member ofthe Trade Unions Commission. His influential book on the law relating totrade unions was based on his memorandum to the report of thatCommission.53 Justice Wightman, notwithstanding his exalted position asa judge of the Queen’s Bench, never lost his innate modesty. He had aprofound knowledge of the law and possessed the essential judicialqualities of patience in listening, discrimination in judging, and clearness inexplaining.54 Justice Coleridge, grand-nephew of the poet, had a distin-guished career, first as an academic and then at the bar, culminating in hisappointment as a Justice of the Court of King’s Bench. Currently there isanother Coleridge, presumably a descendant, sitting as a judge of the HighCourt in England. What he thinks of his ancestor’s dissent in Lumley v GyeI do not know. It might be interesting to find out.

50 Holdsworth, above n 16, at 506. See also ibid, at 505–508.51 See, eg, Collen v Wright (1857) 8 E & B 647, 120 ER 241; Indemaur v Dames (1866)

LR 1 CP 274; Barwick v English Joint Stock Bank (1867) LR, 2 Ex 259; Austin v Dowling(1870) LR 5 CP 534.

52 W Holdsworth, History of English Law (London, Methuen & Co Ltd, 1952) vol 13,437. His judgment in Lumley v Gye, however, makes this assertion questionable.

53 Holdsworth, above n 16, at 454.54 E Foss, The Judges of England (New York, AMS Press Inc, 1966) vol 9, 203.






10

Contracting Out of Liability forDeceit, Inadvertent

Misrepresentation and NegligentMisstatement

MARK P GERGEN

THIS ARTICLE EXAMINES when an actor may contract out ofliability for misleading another. I take a broad perspective, coveringcases in which an actor negligently or innocently misleads another

as well as cases in which an actor knowingly misleads another. Myprincipal focus is on the ability of an actor to disclaim liability for amisrepresentation made to induce a contract. I also will look at the abilityof an actor to disclaim liability for misleading a claimant in the claimant’sdealings with a third person. The topic straddles the law of contract, tort,unjust enrichment and equity. In tort, it encompasses the tort of fraud ordeceit and the tort of negligent misstatement.1 There are two limitations onscope. I exclude cases in which claimants suffer bodily harm or physicalharm to their own property as a result of being misled. And I exclude casesin which an actor harms a claimant by misleading a third person. I focuson United States law.

While the law is unsettled on a few important issues, some basic pointsare fairly well established. An actor generally is strictly responsible for theaccuracy of a representation made to induce a contract. To avoid respon-sibility, an actor must warn the other expressly not to rely or, if not that,then disclaim liability expressly should a representation turn out to beinaccurate. An exculpatory agreement will absolve an actor from liabilityfor an inadvertent misrepresentation, including both an innocent and a

1 Americans call it the tort of negligent misrepresentation. I will follow the practice in therest of the common law world of calling the tort negligent misstatement, reserving the termmisrepresentation for a misstatement regarding a contract between the actor and theclaimant.



negligent misrepresentation (though there is some confusion on negli-gence).2 There is significant disagreement over when, if ever, and why anexculpatory agreement will preclude a fraud claim. Part I examines thisissue in some detail. It assesses the reasons for enforcing exculpatoryagreements and concludes that the arguably valid reasons, which generallyinvolve protecting people from baseless accusations of fraud, should beadequately served by a rule that requires fraud to be proved by clear andconvincing evidence.

Turning from misrepresentation to misstatement, the key point is thatgenerally an actor is not legally responsible for the accuracy of informa-tion which her or she disseminates when the information is not suppliedas an inducement to contract or other transaction benefiting the actor,3

though an actor can never knowingly mislead others.4 Generally, an actorwho disseminates information has a duty of care to a recipient only if heror she appears to intend to invite the recipient to rely on the informationin making a decision. The requirement for a duty of care of invitedreliance gives people the ability to define when they have a duty of carein supplying information, and the content and scope of this duty. On theother hand, if an actor does invite a recipient to rely on information, thenan exculpatory agreement will absolve the actor from negligence liabilityonly if the exculpatory term is clear and reasonable in the circumstances.5

2 See below Part II.3 I do not examine the ability of people to absolve themselves from liability for

misleading a claimant in the claimant’s non-contractual dealings with them. Misrepresenta-tion is a legal basis for reversing a gift but rarely will donees attempt to absolve themselvesfrom responsibility for misleading a donor. The problem may arise in the law of equitableestoppel, which will prevent people from asserting a right, claim or defence if it would beinequitable for them to do so given that they misled the claimant to believe they would notassert the right, claim or defence. I explain the relationship of this body of law to cognatedoctrines in the law of contract, tort, and unjust enrichment in MP Gergen, ‘TowardsUnderstanding Equitable Estoppel’ in C Rickett and R Grantham (eds), Structure andJustification in Private Law: Essays for Peter Birks (Portland, Hart Publishing, 2008).

4 Liability for deceit is not limited to misrepresentation inducing a contract with the actor.There is little authority on the power of people to absolve themselves from liability fordeceiving a claimant when they supply information to a claimant regarding a transaction witha third person. The issue will arise only when an actor supplies information pursuant to acontract with a claimant or in a form that can contain exculpatory terms.

5 See below Part III.

238 Mark P Gergen



I. CONTRACTING OUT OF LIABILITY FOR DECEIT

There is an old saying that ‘fraud vitiates all contracts’.6 This is not strictlytrue. The right not to be deceived must be alienable or waiveable.7 Peopleroutinely submit to the risk of being deceived when they play games thatinvolve deception. There may be a few situations in which deception is anaccepted part of commerce. A liar’s loan may be a recent example.8 Andthere are less exotic situations in which it is in a person’s interest to waivethe right to legal redress for fraud in the inducement of a contract. Intheory, this is possible whenever the expected value of having this right isless than the expected cost to the other party to a contract. If it may be ina person’s interest to waive the right to legal redress for fraud, and aperson knowingly agrees to a waiver, then it is difficult to make a case fornot enforcing the waiver. I will return to this point below when I examineconsent as a possible basis for enforcing an exculpatory agreement.

A variation on the old saying gets us close to its true meaning. Thevariation is that ‘no form of contract can stand, if induced by fraud’.9 Thereason is that

the ingenuity of draftsmen is sure to keep pace with the demands of wrongdoers,and if a deliberate fraud may be shielded by a clause in a contract that thewriting contains every representation made by way of inducement, or thatutterances shown to be untrue were not an inducement to the agreement, sellersof bogus securities may defraud the public with impunity, through the simple

6 Wu v Chang 823 A 2d 1197 (Conn 2003); Bogosian v Bederman 823 A 2d 1117 (RI2003); Snyder v Lovercheck 992 P 2d 1079 (Wyo 1999); Albany Urology Clinic, PC vCleveland 528 SE 2d 777 (Ga 2000) (in a dissenting opinion).

7 Whalen v Connelly 545 NW 2d 284, 294 (Iowa 1996) is an uncontroversial example.The claimant learned of the possibility that he had been defrauded in entering into apartnership agreement but agreed to enter into a renewed agreement in which he explicitlydisaffirmed any fraud claim. A more commonplace and controversial example involves theeffectiveness of an incontestability clause in an insurance contract. An incontestability clauseprevents an insurer from denying coverage based on a misrepresentation in procuring theinsurance after the insurance has been in effect for a sufficient period. Many states have lawsrequiring insurance contracts include incontestability clauses. Courts are split on whether ornot the clause precludes a fraud claim. Paul Revere Life Insurance Co v Haas 644 A 2d 1098,1108 (NJ 1994) holds that the clause does preclude denial of coverage based on fraud andstates this is the majority rule. For the contrary view see Mutual Life Insurance Co of NewYork v Insurance Commissioner for State of Maryland 723 A 2d 891, 896 (Md 1999); Estateof Doe v Paul Revere Insurance Group 948 P 2d 1103, 1114 (Hawaii 1997); New EnglandMutual Life Insurance Co v Doe 688 NYS 2d 459, 462 (NY 1999).

8 These are also called ‘stated income’ or ‘no doc’ loans. A loan applicant is told that onemust state an amount or source of income or an amount of assets to qualify for a loan butthat little or no documentation need be supplied. It was also understood the lender would nottry to verify the statement.

9 Arnold v National Aniline & Chemical Co 20 F 2d 364, 369 (2nd Cir 1927) (New Yorklaw).

Contracting Out of Liability 239



expedient of placing such a clause in the prospectus which they put out, or in thecontracts which their dupes are asked to sign.10

There is a narrow proposition here that is uncontroversial. It is that anexculpatory term—for example, a disclaimer, a merger provision, or anon-reliance clause—will not shield fraud if it is in a document which aclaimant is unlikely to have read or to have understood. Courts thatenforce exculpatory agreements exclude terms in form contracts or boiler-plate.11 A related qualification that is universally recognised is that anexculpatory term cannot shield a contract from a challenge on the groundthat consent to the term was obtained by fraud or trickery.12 Some courtsthat enforce exculpatory agreements impose further limitations. Somerequire that the claimant had read the contract and had been representedby a lawyer.13 Some hold a merger provision will preclude a fraud claim onan extra-contractual promise but not a fraud claim on an extra-contractualmisrepresentation of fact.14 Some hold a disclaimer will not shield an actorfrom a fraud claim on a representation involving ‘facts . . . peculiarlywithin the knowledge’ of the actor.15

Danann Realty Corp v Harris16 illustrates the sort of case in which theissue of the enforceability of an exculpatory agreement is up for grabs.17 Itis the leading New York case on the subject. The claimant purchased along-term lease allegedly relying on oral representations by the sellerregarding the property’s operating expenses and profits. The contractincluded terms stating the seller had made no representations regardingexpenses, the buyer had made a personal inspection, the contract was the

10 Ibid.11 Danann Realty Corp v Harris, 157 NE 2d 597 (1959) [Danann Realty] limited the rule

to exclude a ‘general and vague merger clause’ to distinguish Sabo v Delman 143 NE 2d 906(1957), a case from two years earlier that held an exculpatory term would not preclude afraud claim. WE Nelson, ‘From Morality to Equality: Judicial Regulation of Business Ethics inNew York, 1920–1980’ (1999) 43 New York Law School Law Review 223, 276–7 attributesthe embrace of exculpatory agreements in Danann Realty to a shift on the court to a majoritythat favoured business efficiency over business ethics. A later New York case holds anegotiated merger clause will also preclude a fraud claim on an extra-contractual representa-tion: Citibank v Plapinger 485 NE 2d 974 (NY 1985).

12 Toy v Metropolitan Life Insurance Co 928 A 2d 186 (Pa 2007); Yocca v PittsburghSteelers Sports Inc 854 A 2d 425 (Pa 2004); Hamade v Sunoco Inc (R & M) 721 NW 2d 233(Mich App 2006).

13 See LaFazia v Howe 575 A 2d 182 (RI 1990) (giving effect to an exculpatory term whenthe claimant read the contract and was represented by a lawyer).

14 Watkins & Son Pet Supplies v Iams Co 254 F 3d 607 (6th Cir 2001) (Ohio law).15 Aetna Cas & Surety Co v Aniero Concrete Co Inc 404 F 3d 566, 575–76 (2nd Cir

2005); Banque Arabe et Internationale D’Investissement v Maryland National Bank 57 F 3d146, 155 (2nd Cir 1995).

16 Above n 11.17 For the position that exculpatory terms cannot preclude a fraud claim, see, eg, Lusk

Corp v Burgess 332 P 2d 493 (Ariz 1958); Aspiazu v Mortimer 82 P 3d 830 (Idaho 2003);First National Bank in Durant v Honey Creek Entertainment Corp 54 P 3d 100 (Okla 2002);Shah v Racetrac Petroleum Co 338 F 3d 557 (6th Cir 2003) (Tennessee law).

240 Mark P Gergen



complete agreement, and neither party was relying on representationsoutside the contract. The New York Court of Appeals held the contractprecluded a fraud claim by treating the exculpatory terms, particularly thenon-reliance clause, as conclusive proof of the absence of fraud. I will comeback to the arguments for enforcing an exculpatory agreement after Iexamine how courts that refuse to enforce exculpatory agreements handlefraud claims.

A. ‘No Form of Contract Can Stand, If Induced by Fraud’

The proposition ‘no form of contract can stand, if induced by fraud’, iftaken literally, would mean it was impossible to write a contract that willpreclude a party from challenging the contract by claiming fraud in theinducement. This might seem to belie my claim earlier that the right not tobe deceived is alienable or waiveable. If it truly were impossible to write acontract to preclude a claim of fraud in the inducement, then the right notto be deceived regarding a contract in effect would be inalienable andunwaiveable.

While this may be the practical effect of the rule, none of the reasons forthe rule requires denying the possibility that a person could agree to bearthe risk of being deceived. The rule is grounded largely on the observationthat people often ‘accept . . . and act upon agreements containing . . .exculpatory clauses . . . but where they do so, [they are] nevertheless, inreliance upon the honesty’ of the actor.18 The explanations for this sort ofbehaviour are familiar. When an exculpatory term is in boilerplate in aform contract it probably is unread. Because of the strong expectation ofhonesty, even if people do read an exculpatory term, they may not think itcovers dishonesty, or they may underestimate the risk of dishonesty.19 Butthese are merely reasons to be sceptical about the quality of assent toexculpatory agreements. Reasons of this type cannot establish that it couldnever be in a person’s interest to agree to bear the risk of being deceived.The other argument for the rule is that the law cannot spell out whenexculpatory terms will and will not be enforced for the cunning will exploit

18 Snyder v Lovercheck, above n 6, at 1086. And, we might add, they would not thinkthey had agreed to relinquish the right to seek legal redress for deceit.

19 JM Lipshaw, ‘Of Fine Lines, Blunt Instruments, and Half-Truths: Business AcquisitionAgreements and the Right to Lie’ (2007) 32 Delaware Journal of Corporate Law 431, makesan argument along these lines to justify a rule construing exculpatory terms narrowly. RPrentice, ‘Contract-Based Defenses in Securities Fraud Litigation: A Behavioral Analysis’[2003] University of Illinois Law Review 337, argues that often when people do agree toexculpatory terms it is because they irrationally underestimate the likelihood that the personwith whom they are dealing is acting dishonestly.




any such rule (or at least any feasible rule20) to defraud the unwitting. Butthis does not foreclose the possibility of enforcing an exculpatory agree-ment. It only forecloses having rules to define when an exculpatoryagreement will be enforceable.

But neither are there strong reasons to reject the rule. The best reason forenforcing exculpatory terms is to protect innocent people from baselessaccusations of fraud. A persistent worry in the United States—attributablein significant part to our reliance on the jury to resolve contestable issuesof fact (particularly when a fact issue turns on credibility), to distrust of thejury, and to the high cost of litigation—is that people who are disappointedwith a contract, but who have no contractual grounds for complaint, willclaim fraud on slight evidence in the hope of finding a sympathetic jury orin the hope that the high cost of litigation will prompt the other party tosettle on favourable terms. Rules requiring that fraud be pled withspecificity and be proven by clear and convincing evidence are safeguardsagainst this sort of behaviour. They enable courts to police fraud claimsand direct courts to resolve factual doubts against a claimant. What anexculpatory agreement will not do is to shield an actor from having todefend himself against a fraud claim if the claimant can make out aplausible case. And an exculpatory agreement will not shield an actor fromliability for fraud if there is strong evidence.

Snyder v Lovercheck shows how the approach works in practice.21

Snyder purchased the Loverchecks’ wheat farm during the winter afterinspecting the farm numerous times. Many of the visits were in thecompany of Snyder’s agent (Hayek) and the seller’s agent (Ron Lover-check). Snyder alleged that on the first visit he was told by Ron that 100 ofthe 1,960 acres had a problem with rye, which reduces wheat production.Later Ron told Snyder that he had confirmed this by talking to a priorowner. The sales agreement, which was a form contract published by thestate real estate commission, said the farm was sold ‘as is’ and that thebuyer was not relying on representations made by the seller or the seller’sagents. When the first crop came in Snyder discovered that the rye’spresence was much worse than he had been told and that it reduced thevalue of the land by one-quarter of the purchase price. He brought suitasserting breach of warranty, negligent misstatement, and fraud. He alsosued his agent, Hayek, asserting negligence for failing to warn him about

20 I can imagine a rule giving effect to an exculpatory term that would be difficult for thecunning to exploit. For example, the law might require an earnest statement from theclaimant to the effect: ‘I know you may lie to me and I agree to bear that risk.’ But it isdifficult to imagine someone asking for such a concession in the usual settings in which it is ina person’s interest to agree to waive the right to legal redress for fraud. One such setting iswhere a person has good reason to trust an actor but recognises others who do business withthe actor may not be as trusting and may falsely accuse the actor of fraud.

21 Above n 6.

242 Mark P Gergen



the exculpatory terms. The trial court granted summary judgment for alldefendants on all claims and awarded costs and legal fees to the sellers andtheir agent. I will return to the court’s handling of some of the other claimslater.

On appeal the Loverchecks and Ron sought to preserve their victory byurging the court to adopt the New York rule enforcing exculpatoryagreements. The court declined to do this. But the court’s rejection of theNew York rule did not lead it to remand the case for a trial on the merits ofthe fraud claim. The court concluded summary judgment was appropriate,stating as a rule ‘fraud will not be imputed to any party when the facts andcircumstances out of which it is alleged to arise are consistent with honestyand purity of intention’.22 In other words, dishonesty will not be imputed ifthere is a reasonable chance an actor did not know a representation wasfalse (and was not reckless with regards to its accuracy), or did not intendor have reason to expect that the claimant would rely on the representa-tion.23

I think the Wyoming court got it right both in rejecting the New Yorkrule and in adopting rules that require resolving doubts about the existenceof fraud against a claimant. The policy strikes a decent balance betweenthe competing goals, which generally are deterring and redressing deceitwhile protecting honest actors from baseless accusations of deceit. Thisalso achieves the purpose of most exculpatory agreements. And byrejecting the New York rule the court avoided the difficulties of definingwhen and explaining why an exculpatory agreement will shield fraud. Iturn to these difficulties now.

B. Enforcing Exculpatory Agreements

Often when courts enforce an exculpatory agreement it is to dismiss aweak claim that could also have been dismissed under rules that requireresolving doubt about the existence of fraud against a claimant. But thereare cases in which an exculpatory agreement seems to have decisive effect.

22 Ibid, at 1086.23 Cushman v Kirby 536 A 2d 550 (1987) illustrates the sort of facts that suffice to

establish deceit in a similar situation. This was a sale of a home ‘as is’ in which the wife of thecouple selling the home responded to an inquiry about the well water ‘It’s good. It’s fine.’ Herhusband stood silently by while she said this. The buyers did not taste the water. When theymoved in they discovered the water had a strong sulphur taste. The buyers were allowed torecover the cost of hooking up to the municipal water supply on grounds of fraud. Thehusband admitted he was aware of the problem and that he heard his wife’s statement. Thejury was allowed to infer that he remained silent hoping to induce the buyers to purchase thehome. As for the wife, the jury was allowed to infer that she knew her statement was partialand misleading and she made it hoping to induce the buyers to purchase the home. Snyder vLovercheck would be similar if the sellers had known of their broker’s representation andthey had known from prior crops that the rye’s presence was much worse than represented.




LaFazia v Howe is an example.24 The Howes, an elderly couple, purchaseda delicatessen from LaFazia and Gasrow. The Howes’ court papers tell aplausible story of deceit. They alleged the sellers told them the deli grossed$450,000 to $500,000 per year. The Howes initially were dubious becausethe seller’s tax returns showed a much lower income. The sellers explainedthis away by pointing to their fancy houses and fancy cars and their lack ofanother source of income. Soon after they took over the deli the Howesdiscovered its income was a fraction of what had been represented. Theytried to make a go of it but after six months they sold the deli for half ofwhat they had paid for it. The contract contained strong terms disclaimingrepresentations and reliance and stating the deli was sold ‘as is’. The trialjudge granted the sellers’ motion for summary judgment. He admonishedthe Howes from the bench that they had seen ‘the tax returns didn’t justifythe asking price’ and that the contract had warned them they were ‘makingtheir own judgment’ and ‘acting upon their own’. The Rhode IslandSupreme Court affirmed, holding that a clear exculpatory term precludes afraud claim. In the rest of this part I examine the arguments for enforcingexculpatory agreements. I conclude they do not justify the result in LaFaziav Howe, and generally do not justify enforcing an exculpatory agreementwhen there is clear and convincing evidence of fraud, with the possibleexception being an agreement absolving an actor from liability for thirdparty fraud.

(i) Some Clearly Bad Arguments

Courts like those in New York and Rhode Island that enforce exculpatoryagreements struggle to square this result with the law of deceit. This isdifficult to do because the law of deceit is designed to protect people likethe Howes, who are naïve, stupid or too trusting for their own good, frompeople like LaFazia and Gasrow. The so-called ‘Double-Liar’ argument isthe worst of the bunch.25 The gist of the argument is that claimants canassert a fraud claim only if they lied in representing that they did not relyon a representation. Or as one judge has said: ‘To allow the buyer toprevail on its claim is to sanction its own fraudulent contract. Theenforcement of non-reliance clauses recognises that parties with free willshould say no rather than lie in a contract.’26

24 Above n 13.25 The argument appears in Danann Realty, above n 11. People who should know better

have made the argument. It is made by Judge (now Justice) Alito in MBIA Insurance Corp vRoyal Indemnity Co 426 F 3d 204, 218 (3rd Cir 2005), by Judge Easterbrook in Rissman vRissman 213 F 3d 381 (7th Cir 2000), and by Vice Chancellor Strine in ABRY Partners vF&W Acquisition, LLC 891 A 2d 1032 (Del Ch 2006). [ABRY Partners].

26 ABRY Partners, ibid, at 1058.

244 Mark P Gergen



The silliness of the argument is clear if we consider its implications inLaFazia v Howe. The argument rests on the dubious factual premise thatthe non-reliance clause was understood by the Howes to be a meaningfulrepresentation of their state of mind in buying the deli. Even conceding thispremise, the representation would not be fraudulent if the sellers knew itwas false—that is, if the sellers knew the Howes were relying on theirrepresentations notwithstanding the non-reliance clause. If the Howes didanything that might be dishonest, then it is by asserting a fraud claim to getaround defences to a contract claim. But it is dishonest for the Howes toclaim fraud only if they knew the contract was meant to absolve the sellersfrom liability for deceit. And even then this would be dishonest only if theHowes were insincere in agreeing to absolve the sellers from liability fordeceit. We would have to imagine that the Howes anticipated that thesellers might be lying, that the Howes understood the contract purportedto absolve the sellers from liability in the event they were lying, and thatthe Howes slyly agreed to the contract secretly intending to sue the sellersfor fraud if it turned out the sellers were lying. People do not behave in thisway.

In rejecting the Howes claim, the trial judge admonished them that theyhad seen the tax returns and that the contract had warned them ‘they wereacting on their own’. Perhaps the trial judge dismissed the claim because hethought the Howes had acted unreasonably. A few cases hold that a clearexculpatory term renders a claimant’s reliance unreasonable, absolving thedefendant from liability for having misled the claimant.27 But this flies inthe face of the long-standing principle that contributory negligence is not adefence to fraud.28 I will come back to the related question of when anexculpatory agreement may render a claimant’s reliance ‘unjustifiable’,which is an element of fraud under United States law.

A recent decision by the Delaware Court of Chancery in ABRY Partnersv F&W Acquisition, LLC29 stakes out a middle ground between theposition that a contract can never shield fraud and the position that a clearexculpatory agreement should be enforced so long as it is not procured byfraud or trickery. Unfortunately, while the court reached a defensibleresult, the opinion, which is by Vice Chancellor Strine, makes a muddle ofthe problem conceptually. The claimant ABRY bought the stock of apublishing company from its owner, a private equity firm, for $500million. The acquisition agreement had a term that is standard in suchagreements stating that all representations were made by the targetcompany and not by the seller. The agreement also limited the seller’sobligation to a $20 million indemnity expressly stating ‘misrepresentation’

27 Masingill v EMC Corp 870 NE 2d 81 (Mass 2007), which I discuss later, is an example.28 See Restatement (Second) of Torts (1977) § 545A, Illustrations 1–3.29 Above n 25.




to be among the covered grounds. After the closing ABRY discovered factsthat led it to believe that the target’s financial information had beenmanipulated to inflate the target’s value by $100 million and that the sellerwas complicit in the manipulation.

The case holds that a contract cannot absolve a seller from liability forfraud for representations found within the contract if the seller knew therepresentations were false. This leaves three exceptions covering situationsin which exculpatory terms may be enforced. All are limited to ‘negotiatedcommercial contracts’ involving ‘sophisticated parties’. First, a clausedisclaiming reliance will absolve the seller from extra-contractual represen-tations (that is, representations not repeated within the contract). This issimilar to the rule in Danaan Realty. Secondly, a contract may absolve anactor from liability for a reckless misrepresentation.30 Thirdly, a contractmay absolve an actor from liability for dishonesty of another person solong as the actor is unaware of the dishonesty. In the situation of ABRYthis meant the seller could absolve itself from liability for fraud by thetarget company.31

Much of this flows from two opposing principles set up by ViceChancellor Strine in the opinion. One principle abhors dishonesty. Thisprinciple is so strong, according to Vice Chancellor Strine, that anagreement absolving an actor from liability for dishonesty is void asagainst public policy.32 The other principle favours private ordering. Thegeneral rule and two of the exceptions follow straightforwardly from thetwo principles. The rules enable people to agree to absolve an actor fromliability for fraud except in situations involving personal dishonesty. Theprinciple favouring private ordering holds until it runs square into theprinciple abhorring dishonesty.

I do not think answers can be found reasoning from these particularprinciples. The two principles do not explain why an exculpatory agree-ment should absolve an actor from liability for a representation notrepeated within a contract (policy reasons are the best explanation forbeing extra-sceptical about alleged oral representations). While there issome merit in the distinction between actors absolving themselves fromliability for personal dishonesty and actors absolving themselves from

30 Fraud usually encompasses a reckless misrepresentation, meaning a representation madewith reckless indifference to its possible falsity. A related concept defines as being fraudulentaverring a fact as true with knowledge there is not adequate basis for ascertaining the truth ofthe fact.

31 The third rule could develop into a significant exception if it is extended to permit aprincipal to absolve himself from liability for an agent’s fraud.

32 ABRY Partners, above no 25, at 1062. Vice Chancellor Strine even suggests such a rulemay be justified on efficiency grounds. If the detection and correction of fraud was costlessand error-free this generally would be true (putting to the side games involving deception). Ofcourse, it is not, and so it is not clear that it could never be in the joint interests of contractingparties to preclude a fraud claim.

246 Mark P Gergen



liability for dishonesty by a third party,33 Vice Chancellor Strine draws toosharp a line. The primary reason to enforce exculpatory terms, which is toprotect honest actors from baseless accusations of fraud, applies to claimsof personal dishonesty. The primary reason not to enforce exculpatoryterms, which is scepticism about the quality of assent, applies to claims ofthird party dishonesty. Indeed, much fraud is committed by agents. Therealso is some merit in the distinction between knowing and recklessmisrepresentation. At a fundamental level inadvertent misrepresentation isdifferent from intentional misrepresentation both as a matter of moralityand as a matter of policy. But intentional, knowing and reckless misrepre-sentation are difficult to distinguish, particularly once one grapples withproblems of proof.

(ii) Policy

Judge Augustus Hand aptly summarised the policy argument for enforcingexculpatory agreements in 1927, ascribing the view to Massachusettscourts (which later recanted):

The Massachusetts cases emphasize the desirability of certainty in the contrac-tual relations of those who have made a definite agreement, and if they say thatthey contract without regard to prior representations and that prior utteranceshave not been an inducement to their consent, any occasional damage to theindividual caused by antecedent fraud is thought to be outweighed by theadvantage of certainty and freedom from attacks, which would in the majorityof cases be unfounded where such provisions were in the agreement.34

The key claim is in the final words. This is a claim that the absence offraud can be inferred from the presence of an exculpatory agreement. Thisclaim is wildly unrealistic if one takes the entire universe of contracts. Butthe rule enforcing exculpatory agreements excludes form contracts andboilerplate and is limited to what may loosely be described as negotiatedcommercial contracts. An exculpatory agreement has some evidentiaryvalue in such a contract. But a rule enforcing an exculpatory agreementgoes further and treats it as conclusive evidence of the non-existence offraud, preventing a court from considering other evidence no matter howstrong it might be. So in Danann Realty the court said ‘a specific disclaimerdestroys the allegations in the complaint that the agreement was executedin reliance upon these contrary representations’. A charitable interpretationof this is that a disclaimer ‘destroys’ reliance by conclusively establishingits non-existence.

33 See below Part I (B)(iv).34 Arnold v National Aniline & Chemical Co, above n 9, at 20. Judge Hand also cites to

25 Columbia Law Review 231, which is a brief case comment.




The argument for this is not entirely evidentiary. American courts andscholars have long argued over the related question whether a mergerprovision in a contract should be treated as conclusive evidence that thecontract was indeed a final and complete statement of the parties’agreement. The view that it should be, which is associated with Williston,is best justified not by the evidentiary value of a merger provision butinstead by other purposes served by treating a merger provision asconclusive. These include encouraging people to put their agreements inwriting, simplifying contract enforcement, making interpretation an issuefor the judge and not the jury, and strengthening the hand of appellatejudges in policing contract interpretation.35

Connecting the evidentiary and institutional arguments are the fears thatpeople over-claim fraud and that courts do a poor job of distinguishinggood claims from bad. If these fears are borne out by reality, then a policyof enforcing exculpatory agreements would permit some fraud to gouncorrected while reducing the costs of administering contracts acrossmany cases, including a significant number of cases in which innocentpeople would incur the cost of defending against unfounded fraud claims,and perhaps even eliminating some outcomes in which innocent peoplewould be unjustly found liable for fraud.

The argument for enforcing exculpatory agreements rests on a verypessimistic view of human litigiousness and of the fallibility of legalinstitutions and a very optimistic view of private ordering through con-tract. But even if one holds these views it is not clear the rule is justified; itis a clumsy tool to discourage unfounded fraud claims and avoid unjustfraud verdicts. The rule requires a court to dismiss a fraud claim no matterhow strong the evidence. The rules requiring that fraud be pled withspecificity and proven by clear and convincing evidence serve the sameends while enabling a court to override an exculpatory agreement if there isstrong evidence of fraud. One has to assume a very bleak view of humanlitigiousness and of the fallibility of legal institutions to justify enforcing anexculpatory agreement on evidentiary and institutional grounds in the faceof clear and convincing evidence of fraud. So on to the next reason.

35 JD Calamari and JM Perillo, ‘A Plea for a Uniform Parol Evidence Rule and Principlesof Contract Interpretation’ (1967) 42 Indiana Law Journal 333 nicely capture this at 341 indescribing the gist of the disagreement between Williston and Corbin: ‘Professor Corbin hasan easy task in demolishing the Willistonian approach. In treating the matter of integration asa question of intent, as Professor Williston purports to do, he shows the absurdity ofexcluding all relevant evidence of intent except the writing itself. But … Williston … [is]unconcerned about the true intention of the parties . . . [Williston is] advocating and applyinga rule of form.’

248 Mark P Gergen



(iii) Unjustifiable Reliance

The actual doctrinal basis for the decision of the Rhode Island SupremeCourt in LaFazia v Howe was that the exculpatory agreement made theHowe’s reliance ‘unjustifiable’. The law of unjustifiable reliance may seeman unpromising place to look for a rationale for enforcing exculpatoryagreements. Dan Dobbs’ treatise concludes that the concept of unjustifiablereliance has no analytical content in the law of fraud. He argues that courtsuse the requirement of justifiable reliance as a tool to screen out claimswhich are weak on some other ground (that is, a weakness in proof of thefact of a misrepresentation, or of a claimant’s reliance, or of an actor’sdishonest intent).36 The requirement of justifiable reliance is worrisome forit is easy to make the mistake of equating justifiable reliance withreasonable reliance. An additional reason for scepticism about the value ofthe concept is that it appears only in United States law. Other legal systemsseem to do fine without it.

The confusion of unjustifiable reliance with unreasonable reliance isunderstandable. Much of the law of unjustifiable reliance is rules definingwhen an actor who may have knowingly or recklessly supplied inaccurateinformation to a claimant is not liable for fraud because the actor probablydid not intend, or have reason to expect, the claimant would rely on theinformation because a reasonable person in the claimant’s situation wouldnot rely on it. The mens rea of fraud requires that an actor intend, or havereason to expect, that a claimant would attach significant importance to arepresentation.37 This is in addition to the requirement that an actor knowa representation is inaccurate or be reckless with regards to its accuracy.The rules precluding liability for opinion and prediction are best explainedon this basis. A salesman who puffs his wares may not intend or havereason to expect a buyer will attach importance to the puffery in decidingwhether to buy because most people discount puffery.

Often the issue is amenable to being resolved by rule. When an actor’sactual intent is unknown, liability will turn on whether the actor hadreason to expect the claimant would attach importance to a representation.This in turn depends on whether a normal or reasonable person would beexpected to do so, unless an actor has particular reason to know a claimant

36 D Dobbs, The Law of Torts (St Paul, West Group, 2000) 1360–61. Dobbs’ diagnosis ofthe law of unjustifiable reliance resonates with the policy argument for enforcing exculpatoryterms just discussed. Bruce Feldthusen concludes the element of justifiable reliance serves nofunction in the law of negligent misstatement that is not as well served either by the test ofpecuniary interest for duty or by the defence of contributory negligence. B Feldthusen,Economic Negligence, 3rd edn (Scarborough, Carswell, 1994) 64, n 166.

37 Restatement (Second) of Torts (1977) § 531 is a clear statement of the general principle.The Comments to the Section focus on the issue of liability to remote and numerousclaimants.




is credulous. Whether a normal person would be expected to act on arepresentation turns on social or general facts, and so the issue is amenableto being resolved by a rule. Again the rule on puffery is an example. Thegeneral rule treating reliance on opinion as unjustifiable is another exam-ple. In United States law most of these rules are found in the law ofunjustifiable reliance. In other common law systems these rules tend to be agloss on a rule restricting misrepresentation to statements of fact. Theserules should give way if there is sufficiently strong evidence that an actorknew, or had reason to expect, a claimant was unusually credulous.

An exculpatory agreement sometimes will negate this aspect of the mensrea of fraud. Masingill v EMC Corp illustrates this point.38 Masingill left ajob at Compaq, relinquishing a bonus and stock options, to work for DataGeneral. Because of moving, she worried about being protected againstlosing stock options in Data General if the company were bought out.Senior executives at Data General were protected against this risk by‘forward vesting’. Mid-level employees like Masingill were not. Masingillsought and received assurances from senior executives at Data General thatthey would work to get her ‘chute’ protection during the first year. But theemployment contract, which was the product of much going back andforth between the parties, guaranteed her only a year’s salary as severancein the event of a buyout leading to her termination. Masingill understoodthat this left her without the ‘chute’; this was the major bone of contentionin negotiating the contract.

Masingill was right to worry. EMC bought out Data General eightmonths into her tenure. She quit four months later when a suitable jobcould not be found at EMC, losing her stock options. She sued DataGeneral and the executives individually alleging they had deceived her inassuring her that efforts would be made to get her the ‘chute’ during thefirst year. The case was tried by a jury, which found that the executives had‘knowingly and recklessly made false statements’ to Massingill aboutgetting her the ‘chute’. Nevertheless a verdict was rendered for theexecutives. The jury was instructed by the trial court that it had to findMassingill’s reliance on the representations unreasonable because therepresentations were explicitly contradicted by her employment contract.The Massachusetts Court of Appeals affirmed, reasoning that the contractmade Massingill’s reliance on the representation unreasonable as a matterof law. This basis for the decision is inconsistent with the general principlethat contributory negligence is not a defence for fraud. A better rationale isthat the executives probably did not intend, and did not have reason toexpect, Massingill would attach importance to their assurances that theywould work to get her the ‘chute’ given the inconsistent terms in the

38 870 NE 2d 81 (Mass 2007). I have simplified the facts to isolate the relevant claim.

250 Mark P Gergen



employment contract and the negotiation history. The unreasonableness ofMassingill’s reliance negates the mens rea of deceit. There is an importantsubstantive distinction between non-culpability for unexpected relianceand contributory negligence. A salesman who knows of and exploits acredulous buyer’s unreasonable reliance on puffery commits fraud becausethe salesman knows of the buyer’s credulousness.

This reasoning cannot justify a general rule enforcing exculpatoryagreements even if the rule is limited to negotiated commercial contracts. Ageneral rule elides the distinction between non-culpability for unexpectedreliance and contributory negligence, if the rule requires a court todisregard evidence that an actor intended, or had reason to expect, that aclaimant would unreasonably rely on a representation despite the exculpa-tory agreement. This does not mean that the trial judge erred in Massingillin directing a verdict for the defendants. The policy argument for enforcingexculpatory agreements justifies a directed verdict once Massingill failed toproduce clear and convincing evidence that the executives were aware shecontinued to rely on their representations they would get her the ‘chute’notwithstanding the contract terms to the contrary.

But this was not the case in LaFazia v Howe, and so this reasoningcannot justify the result in that case If you believe the Howes’ allegationthat the sellers explained away the deli’s low reported earnings by implyingthey cheated on their taxes (and the trial judge did, for he stressed this factin dismissing the claim), then this is clear and convincing evidence that thesellers wanted the Howes to rely on the (lying) representations of theearnings. Why else would the sellers embarrass themselves by implyingthey were tax cheats? The Howes’ continued reliance on the representa-tions may have been unreasonable, but that is irrelevant if the sellersintended them to rely. We still do not have a good reason for enforcing anexculpatory agreement when there is clear and convincing evidence offraud (including evidence that the defendant intended the claimant toattach importance to the representation). So we move on to the last reason.

(iv) Consent

Consent is generally not offered as a rationale for enforcing an exculpatoryagreement. ABRY Partners v F&W Acquisition, LLC goes so far as toreject consent as a basis for enforcing an exculpatory agreement.39 Thecontract included a $20 million indemnity cap that expressly covereddamages resulting from ‘misrepresentation’. ABRY argued that the cap wasmeant to cover innocent or negligent misrepresentation but not fraud andthat the cap was meant to cover claims in contract but not tort. Vice

39 Above n 25.




Chancellor Strine properly rejected these arguments as undercut by com-mon usage. The term ‘misrepresentation’ usually refers to fraud or deceitand to the tort action. But Vice Chancellor Strand refused to hold ABRY toits contract. Instead he held that an exculpatory agreement was void asagainst public policy except in limited situations.40

I can see no obvious reason in principle why consent cannot be a basisfor enforcing an exculpatory agreement. As I explained earlier, the reasonsfor not enforcing exculpatory agreements are the suspect quality of consentand the inadvisability for having a rule defining when an exculpatoryagreement is enforceable. These could be sufficient reasons to adopt ageneral rule of invalidity, if it never or rarely was in a person’s interest towaive the right redress for fraud in the inducement of a contract. But thisseems unlikely.

Vice Chancellor Strand identifies one counter-example in ABRY Part-ners. Sometimes it is very costly for a company selling a division to verifythe honesty of the division’s managers or the accuracy of its financialinformation. In this situation it makes a great deal of sense to absolve theseller from liability for fraud in the division. This puts the buyer on noticeto be wary of fraud in the division and avoids the cost of duplicativeanti-fraud investigations. More generally, an exculpatory agreement maybe efficient41 in a situation involving third party fraud in which an actordoes not have a comparative advantage in preventing the third party fraudor in bearing the risk of the third party fraud. An exculpatory agreementmay be efficient even if an actor has a comparative advantage if a claimantwould take duplicative precautions in any event because the claimantcannot rely on fully recovering a loss on third party fraud from the actor.

The assumptions underpinning the policy argument for enforcing excul-patory agreements suggest a more general counter-example. The assump-tions are, first, that often fraud claims are unfounded and, secondly, thatcourts do a poor job of distinguishing good claims from bad. The situationis that Albert has good reason to depend on Bill’s honesty but A knows Bdeals with other people who do not trust B and who can be expected tobring an unfounded claim of fraud if they are sufficiently unhappy with theoutcome of a contract. A may get a lower price by waiving his right to

40 Restatement (Second) of Contracts (1981) § 196 seems to agree. It states: ‘A termunreasonably exempting a party from the legal consequences of a misrepresentation isunenforceable on grounds of public policy.’ The Comments do not explain what, if anything,would be a reasonable exculpatory agreement, though they do suggest a term ‘reasonably’limiting the time in which misrepresentation can be asserted is enforceable. The prohibition issomewhat undercut by a statement that the rule does not apply ‘to language that prevents themaking of any misrepresentation in the first place, such as that disclosing the truth (see §161). Nor does it apply to language that prevents reliance by the recipient on a misrepresen-tation (see § 167) or that makes his reliance unjustified (see § 172), but such language is noteffective unless it actually has the asserted effect and is not a mere recital that it does.’

41 By this I mean that it increases the joint expected return on a contract.

252 Mark P Gergen



redress for fraud in the inducement without reducing the expected value ofthe contract to him.42 More generally, an exculpatory agreement may beefficient if the expected value to the claimant of having the option to seekredress for fraud in the inducement is less than the expected cost to theactor.

The consent argument has some conceptual advantages over the otherarguments for enforcing exculpatory agreements. It avoids tendentiousarguments that an exculpatory agreement negates the fact of fraud. It isconsistent with the practice of making the effect of an exculpatoryagreement an issue for the judge with the key questions being the validityof the claimant’s consent to the exculpatory terms and the apparentmeaning of the terms. It resonates with the policy argument but distancesus from some of its troublesome factual and normative assumptions. Wecan be agnostic about human litigiousness, the fallibility of legal institu-tions, and the relative weight to be assigned to the good and bad effects ofenforcing exculpatory agreements, and defer to private choices. All weneed is faith in private ordering.

Faith in private ordering is the sticking point, of course. Scepticismabout the quality of consent to an agreement absolving an actor fromliability for fraud remains. This scepticism requires two and maybe morelimitations on a rule enforcing an exculpatory agreement. First, this islimited to negotiated commercial contracts. Secondly, there must be aplausible explanation as to why in the situation it was in the claimant’sinterest to waive the right to redress for fraud. If one is sceptical about thequality of consent to exculpatory agreements in general, or to the qualityof consent in the particular case, then one would require more. One mightenforce an exculpatory agreement only if claimants almost surely under-stood they were waiving the right to redress for fraud. And, other than incases of third party fraud, one might refuse to enforce an exculpatoryagreement when there is clear and convincing evidence of fraud on thereasoning that the likely purpose of an exculpatory agreement is served byshielding an actor from unfounded claims of fraud. If you come out here,then you are close to the position associated with the precept ‘No form ofcontract can stand, if induced by fraud’, but you would recognise a narrow

42 Consider a stylised example. Sam (S) is a seller of goods that may be of low or highvalue. The actual value of a good is unobservable to a buyer. The actual value sometimes isknown to S but S’s lack of such knowledge is unverifiable. S has a well-deserved reputationfor honesty. He faces buyers who differ in their willingness to trust S and in their litigiousness.S is unable to differentiate among them. S always discloses that a good is of low value if heknows it, but he knows that if he does not identify a good as being of low value he faces a riskthat a buyer who does not trust him and who is litigious will successfully claim that he knewand failed to disclose a good was of low value. In this situation, buyers who trust S and/orwho are not litigious may obtain a lower price by precommitting not to sue for fraudulentinducement.




exception (for third party fraud) and you would be resolute aboutresolving factual doubts against a claimant, particularly when there is aclear exculpatory agreement in a negotiated commercial contract.

II. CONTRACTING OUT OF LIABILITY FOR AN INNOCENT OR ANEGLIGENT MISREPRESENTATION REGARDING A CONTRACT

It is basic contract law that absent fraud actors may absolve themselvesfrom liability for a representation regarding the subject matter of acontract by getting the claimant to agree that the representation is not partof the contract or getting the claimant to agree that claimants bear the riskof the inaccuracy of a representation. When the representation is not aterm of the contract,43 this is done by including a disclaimer, a non-relianceclause, or a merger provision in a written agreement. In the United States,the parol evidence rule is the principal legal vehicle for giving effect to suchan agreement. If a court applies the strong form of the parol evidencerule—favoured by Williston and the First Restatement of Contracts44—then a clear exculpatory term will absolve an actor from liability for anextra-contractual representation no matter how strong the proof of therepresentation.45 If a court applies the weak form of the rule—favoured byCorbin, the Second Restatement of Contracts, and the UCC—then anexculpatory term will not absolve an actor from liability for an extra-contractual representation if there is credible evidence of the representationand the court finds that the claimant reasonably believed the representa-tion was part of the contract despite the exculpatory term.46

43 When a representation is a term of a contract this is done by a limitation of remedy orsome other term assigning the risk of the inaccuracy of the representation to the claimant.

44 A good summary of the different forms of the rule may be found in LA Cunningham,‘Toward a Prudential and Credibility-Centered Parol Evidence Rule’ (2000) 68 University ofCincinnati Law Review 269.

45 Mitchill v Lath 160 NE 646 (NY 1928) illustrates this. The trial court found the sellerof a house had orally promised to move an unsightly ice house he owned on adjoiningproperty. The majority held the promise was unenforceable because it was omitted from thewritten contract when normally one would expect it to be included. The majority commentedon the seller’s ‘moral delinquencies’ in not fulfilling its promise and observed: ‘[We] havebelieved that the purpose behind the rule was a wise one not easily to be abandoned.Notwithstanding injustice here and there, on the whole it works for good.’

46 Husky Spray Service Inc v Patzer 471 NW 2d 146 (SD 1991) illustrates this. The sellerorally represented to pilots representing the buyer who test flew a used airplane that the planewas ‘ready to go’ and that he would repair any defects. Apparently the seller was unaware ofa growing crack in a crankshaft that hampered the buyer’s use of the plane and eventuallyruined the engine. The sales agreement disclaimed all warranties, said the plane was beingsold ‘as is’, and that the buyer was relying on its own inspection. Recovery was allowed on atheory of breach of warranty. The warranties were established by credible evidence (two ofthe pilots no longer worked for the buyer) and the seller had not specifically bargained for thedisclaimers (they were on the reverse side of the contract, which the buyer testified it neverread).

254 Mark P Gergen



The differences between these rules and the rules on fraud are striking.Even the weak form of the rule assumes that an actor may disclaim liabilityfor an extra-contractual representation regarding a contract by getting theother party to agree that the representation is not part of the contract. Italso assumes that an apparent agreement will suffice. Claimants will loseeven though they believe a representation is actionable if this belief isunreasonable in the circumstances. The difference between the weak formand the strong form of the rule is that under the strong form of the rule aclear exculpatory term is conclusive on the issue of apparent assent whileunder the weak form of the rule a court considers other facts andcircumstances and decides whether, taking account of all of the evidence,the claimant reasonably should have understood that the representationwas not part of the contract.

As for which rule is better, this comes down to one’s views on humanlitigiousness, the fallibility of judges, the institutional considerations thatjustify the strong rule, and, at bottom, whether one cares more aboutfairness in particular cases or generally reducing the cost of contracting. Iwill not try to resolve this conundrum. My interest in this part is moreprosaic. It is with how two bodies of law that provide redress forinadvertent misrepresentation outside of contract law deal with the sameproblem. One is a claim for rescission and restitution on the basis ofinnocent misrepresentation. The other is the tort action for negligentmisstatement. It should be unsurprising that they generally solve theproblem the same way. What is surprising is the amount of confusion theproblem has created, particularly in the law of negligent misstatement. Thehandling of inadvertent misrepresentation as problems of tort and equity,and not as problems of contract, is the source of this unnecessaryconfusion.

A. Innocent Misrepresentation

The doctrine of equitable rescission permits a claimant to rescind acontract for misrepresentation, if the claimant relied on a misrepresenta-tion in entering into the contract and the actor made the misrepresentationfor the purpose of inducing the claimant to enter into the contract.47

Rescission is available even if it was reasonable for the actor to believe thatrepresentation was accurate. Restitution also is available unless interveningcircumstances make this impractical or unfair to either party or to a thirdparty. In addition, in many American states a damage action is available in

47 Hyler v Garner 548 NW 2d 864 (Iowa 1996) (‘the elements of an equitable claim forrescission based on misrepresentation are (1) a representation, (2) falsity, (3) materiality, (4)an intent to induce the other to act or refrain from acting, and (5) justifiable reliance.’)




tort for the difference between value given and value received in anexchange generally measured on the date of the exchange.48 This approxi-mates the value of rescission and restitution on the date of the exchange.

It is fairly well established that an exculpatory term will bar equitablerescission and restitution if it would bar a contract claim on a representa-tion.49 Some courts reach this result by applying the parol evidence rule topreclude proof of a representation.50 Some hold that an exculpatory termnegates reliance.51 The logic of this position seems impeccable. The reasonsthat justify enforcing an exculpatory agreement to shield a party fromliability for contract damages when a representation turns out to beinaccurate as well justify holding the claimant to a bad bargain by denyingrescission and restitution. If the representation appears in the contract,then the reason presumably is that the contract allocated the risk of theinaccuracy of the representation to the claimant. If the representation isextra-contractual, then there are also evidentiary and institutional reasons.

48 Restatement (Second) of Torts (1977) § 552C so provides. The decision to includeinnocent misrepresentation in the Restatement (Second) of Torts was controversial at thetime. Alfred Hill argued in ‘Damages for Innocent Misrepresentation’ (1973) 73 ColumbiaLaw Review 679, and ‘Breach of Contract as a Tort’ (1974) 74 Columbia Law Review 40,that a claim for damages for innocent misrepresentation is best handled in contract law wherethe parol evidence rule can be used to filter out weak claims. Hill also thought the damageclaim too similar to a breach of warranty claim.

FV Harper, F James and OS Gray, The Law of Torts, 2nd edn (Boston, Little, Brown andCo, 1986) § 7.7, respond to Hill. They argue that explicit recognition of innocentmisrepresentation as a basis for reversing a transaction makes the law more coherent andclear as the same result had been achieved through a ‘confusing patchwork’ of mostly‘procedural vehicles’. They also identify a practical reason for imposing strict liability formisrepresentation inducing a contract: the ‘misrepresenter stands to gain at the expense of theother party to the transaction induced by the misrepresentation’. Ibid, at 416. More than aninstinctive belief that it is wrong to profit from another’s misfortune justifies strict liability inthese cases. Strict liability for misrepresentation counteracts the incentive to speak carelesslywhen error benefits a speaker without requiring individualised determinations of fault.

49 See, eg, Wilkinson v Carpenter 554 P 2d 512 (Or 1976) (seller innocently butmistakenly represented roof was in good condition, the contract stated the property was sold‘as is’ with no warranties or representations regarding its quality); Gibson v Capano 699 A 2d68 (Conn 1997) (home seller innocently misstated that termites had not been treated withChlordane while disclosing extent of termite damage he was aware of, buyers knew of termiteproblem but not its extent, contract disclaimed reliance on representations); Creamer vHelferstay 448 A 2d 332 (Md 1982) (holding parol evidence rule precludes rescission basedon a misrepresentation that is inconsistent with the contract); Hoover v Hegewald 689 P 2d965 (Or App 1984) (misrepresentation by broker of number of cattle ranch would supportand of number of irrigated and irrigable acres, contract disclaimed representations by sellerand its agents and stated that the buyers were relying on own investigation).

50 Conflicting secondary authority on the point is reviewed in Wilkinson v Carpenter, ibid,which holds that the rule does apply and that disclaimer precludes redress for innocentmisrepresentation. Restatement (Second) of Contracts (1981) § 196, Comment b, is in accord.For the contrary view that a merger provision or disclaimer does not preclude rescission forinnocent misrepresentation see Halpert v Rosenthal 267 A 2d 730 (RI 1970).

51 Gibson v Capano, above n 49.

256 Mark P Gergen



Nevertheless, a handful of cases suggest there are a limited number ofsituations in which rescission is available for a representation whencontract damages are not available. Britton v Parkin is an example.52 Areal estate broker mistakenly described property as being zoned commer-cial in a real estate listing, advertisements and signage. The buyer wasallowed to rescind the purchase on the basis of innocent misrepresentationeven though the standard form purchase agreement had a merger provisionand boilerplate stating the sale was subject to zoning restrictions. Thebuyer was not allowed to recover expectation damages. A case can bemade for the result, particularly if you side with Corbin and againstWilliston, preferring the weak to the strong form of the parol evidencerule. The result is fair as no one thought there was any doubt about theproperty’s zoning. Allowing rescission and restitution, but not expectationdamages, prevents the seller from reaping a windfall from a mutualmistake while not giving the buyer a windfall if the price was a bargain hadit been zoned as represented. Of course, allowing such claims sacrificessome of the institutional and evidentiary benefits of the strong form of theparol evidence rule. But the unusual facts in these cases (the misrepresen-tation and reliance are indisputable), the limited remedy, and the obscurityof the cases limits the sacrifice.

B. Negligent Misstatement

American courts have struggled with the question whether an exculpatoryterm, such as a disclaimer or a merger provision, will preclude a negligenceaction for a misrepresentation regarding the subject matter of a contract.Some courts avoid the problem by defining the general scope of the tort ofnegligent misstatement in a way that precludes an action based on amisrepresentation regarding the actor’s contract with the claimant.53 Othercourts have adopted a rule that a valid disclaimer or merger provision will

52 438 NW 2d 919 (Mich App 1989). Norton v Poplos 443 A 2d 1 (Del 1982) is similar.An advertisement and sign mistakenly listed property as zoned ‘M-1.’ The buyer was allowedto rescind though a form contract had general merger clause and boilerplate stating thatproperty was subject to restrictions of record. See also Parkhill v Fuselier 632 P 2d 1132(Mont 1981) (listing misrepresented that one acre property had community water supply,form contract said buyer was relying on own investigation and disclaimed oral representa-tions); Lesher v Strid 996 P 2d 988 (Or App 2000) (seller gave buyer a map incorrectlyindicating four acres had water rights, contract had ‘as is’ clause but also referred to transferof water rights).

53 Examples include a rule that a duty of care is owed only when an actor supplies aclaimant with information to guide the claimant in a business transaction with another,National Can Corp v Whittaker Corp 505 F Supp 147 (ND Ill 1981) (Illinois law); a rule thata duty of care is owed only when an actor is in the business of supplying the information,Alderson v Rockwell International Corp 561 NW 2d 34 (Iowa 1997); and the view that theeconomic loss rule bars any form of a negligence action between the parties to the contract




bar the tort action as well as a contract action.54 But some courts havelikened negligent misstatement to fraud and held that an exculpatory termwill not preclude the action. Approaches vary. Some courts hold that amerger provision or general language of disclaimer will not preclude aclaim for negligent misstatement.55 Other courts go further and hold thatcontract terms that purport to absolve an actor from liability for arepresentation have no bearing in a negligent misstatement action otherthan as facts to be considered by the jury in resolving the tort claim.56

Under New York law a tort action is available for a misstatementregarding a contract notwithstanding an exculpatory term if the parties are

when the claim relates to the contract’s subject matter, see Duquesne Light Co v Westing-house Electric Corp 66 F 3d 604 (3d Cir 1995) (Pennsylvania law); Apollo Group Inc v AvnetInc 58 F 3d 477 (9th Cir 1995) (Arizona law); Pulte Home Corp v Osmose Wood PreservingInc 60 F 3d 734 (11th Cir 1995) (Florida law) (the Florida Supreme Court later repudiatedthis version of the economic loss rule); Bailey Farms Inc v NOR-AM Chemical Co 27 F 3d188 (6th Cir 1994) (Michigan law); Danforth v Acorn Structures Inc 608 A 2d 1194 (Del1992); Sebago Inc v Beazer East Inc 18 F Supp 2d 70 (D Mass 1998) (limiting the rule tosales of goods).

54 Sound Techniques Inc v Hoffman 737 NE 2d 920 (Mass App 2000); Rio GrandeJewelers v Data General Corp 689 P 2d 1269 (NM 1984); Stanley v Miro 540 A 2d 1123 (Me1988); Snyder v Lovercheck, above n 6; Brogan v Mitchell International Inc 692 NE 2d 276(Ill 1998); Hodgkins v New England Telephone Co 82 F 3d 1226 (1st Cir 1996) (Maine law);Vermont Plastics Inc v Brine Inc 79 F 3d 272 (2d Cir 1996) (Vermont law); Lowe vAmeriGas Inc 52 F Supp 2d 349 (D Conn 1999), affirmed 208 F 3d 203 (2d Cir 2000)(Connecticut law).

55 See Agristor Leasing v AO Smith Harvestore Products 869 F 2d 264 (6th Cir 1989)(Tennessee law); Keller v AO Smith Harvestore Prods Inc 819 P 2d 69 (Colo 1991);Greenfield v Heckenbach 797 A 2d 63 (Md App 2002); Robinson v Tripco Investment Inc 21P 3d 219 (Utah App 2000); Grube v Daun 496 NW 2d 106 (Wis 1992). Brooks v TimberlineTours Inc 127 F 3d 1273 (10th Cir 1997) (Colorado law) is authority for the corollaryproposition that a clear exculpatory term will preclude a negligent misstatement claim. Thetwo leading cases involve representations by a manufacturer in advertising that an innovativegrain storage silo would better preserve the nutritional value of silage by limiting airexposure: Agristor Leasing v AO Smith Harvestore Products; Keller v AO Smith HarvestoreProds. It turned out that the silos significantly reduced the nutritional value of silage by acombination of air exposure and heat. The claimants did not realise the silo was to blameuntil they suffered a severe loss of milk production and in one case the deaths of several cows.The manufacturer attempted to avoid liability by invoking contract terms that stated thebuyer understood that representations in the advertising were not guarantees and that thebuyer did not rely on the representations. While other grounds for recovery were available(including fraud) the claimants obtained a jury verdict on a claim of negligent misstatement.The court of appeals affirmed announcing the rule stated in text. Horn v AO Smith Corp 50F 3d 1365, 1368 (7th Cir 1995) collects other cases involving claims against Harverstore.Often claimants recovered on a fraud claim.

56 Some cases lump a fraud claim with a negligent misstatement claim to hold that liabilitycannot be avoided by contract. See Moffatt Enters Inc v Borden Inc 807 F 2d 1169 (3d Cir1986) (Pennsylvania law); Gibb v Citicorp Mortg Inc 518 NW 2d 910 (Neb 1994); Gillilandv Elmwood Properties 391 SE 2d 577 (SC 1990). Wilburn v Stewart 794 P 2d 1197 (NM1990) lumps these theories with innocent misrepresentation, albeit in dicta Pearson vSimmonds Precision Products Inc 624 A 2d 1134 (Vt 1993) holds that it is a question for thejury whether the contract was adequate disclosure. Formento v Encanto Bus Park 744 P 2d22 (Ariz App 1987) holds that the parol evidence rule does not apply to a tort claim.

258 Mark P Gergen



in a ‘special relationship’, which is treated as a question for the jurywhenever reasonable people could differ on the character of the relation-ship.57

The New York position is bizarre. New York is among the states thatwill enforce an exculpatory term to preclude a fraud claim. New York alsotakes the Williston approach, enforcing a ‘hard’ version of the parolevidence rule. It is impossible to square this with allowing a claimant to getaround a contractual disclaimer by claiming a representation was negligentand that the parties were in a ‘special relationship’. If the allegedmisrepresentation does not appear in the contract and is disputed, then theevidentiary and institutional arguments for making an exculpatory termconclusive evidence of the non-existence of a representation or of relianceare no weaker when an actor allegedly was negligent regarding theaccuracy of the representation. If a misrepresentation is a term of acontract or is undisputed, and the claimant sues in tort for negligentmisstatement to get around a contract term limiting the right to recover orto sue on the representation, there is no obvious reason for disregarding acontract term allocating the risk of the inaccuracy of a representation tothe claimant, merely because we add the fact that the actor was negligentwith regards to the accuracy of the representation. Generally, a person whowants to ensure the other party to a contract takes reasonable care toensure the accuracy of a representation made as an inducement to contractwill be better off bargaining for strict liability.

The charitable explanation for the New York position is that the court isusing negligent misstatement to move away from the ‘hard’ version of theparol evidence rule and away from its policy of enforcing exculpatory

57 As a consequence in New York claimants have had atypical success getting past motionsfor summary judgment with tort claims for negligent misstatement in contract negotiations.See, eg, Fresh Direct LLC v Blue Martini Software Inc 776 NYS 2d 301 (NYAD 2004)(assurances by provider of software regarding its capacity); CooperVision Inc v IntekIntegration Technologies Inc 794 NYS 2d 812 (NY Sup 2005) (software license and serviceagreement); Fleet Bank v Pine Knoll Corp 736 NYS 2d 737 (NYAD 2002) (assurances byagent of lender that additional financing would be approved; jury question whether there wasa special relationship); Grammer v Turits 706 NYS 2d 453 (NYAD 2000) (broker did notdisclose construction on property adjacent to one-month vacation sublease). New York courtssometimes reject the claim by characterising the parties’ relationship in terms that make onecringe. See, eg, Morris v Putnam Berkley Inc 687 NYS 2d 139 (NYAD 1999) (holding thatemployee is not in a ‘special relationship’ with employer). Oregon has avoided these problemswhile also applying a general rule imposing a duty of care in supplying information when theparties are in a ‘special relationship.’ Oregon has a per se rule precluding a duty of care insupplying information in ‘arms-length negotiations’. Onita Pacific Corp v Bronson 843 P 2d890 (Or 1992). Further, Oregon law defines a ‘special relationship’ as a fiduciary-likerelationship in which an actor undertakes to exercise discretion on the claimant’s behalf. Andunder Oregon law whether there is a special relationship is for the court to resolve. Conwayv Pacific University 924 P 2d 818 (Or 1996) (rejecting claim by professor denied tenure thatuniversity negligently misled him regarding effect of poor student evaluations on tenureprospects).




agreements to preclude fraud claims. The limitation of the negligence claimto ‘special relationships’, a peculiar feature of New York law, limits thisliberalisation of the rules allowing redress on extra-contractual misrepre-sentations on a plausible dimension. An oddity of employing the doctrineof negligent misstatement for this purpose is that the actor’s negligenceregarding the accuracy of an extra-contractual representation becomes anissue. When claimants sue on extra-contractual representations the typicalissues are whether the representation was made as an inducement tocontract and whether the claimants should have understood that they borethe risk of the inaccuracy of the representation. Whether the other partywas negligent regarding the accuracy of the representation is beside thepoint.

A less charitable explanation for the New York position is that the courthas fetishised the classification of negligent misstatement as a tort. The tortaction for negligent misstatement dates back to the 1920s in the UnitedStates58 and to the 1960s in the Commonwealth.59 In both the UnitedStates and the Commonwealth, the action was characterised as a tort, andnot as a claim for negligent performance of a contractual undertaking,because the defendant supplied misleading information to the claimantgratuitously or pursuant to a contract with a third person. Contract wasunavailable because of doctrinal impediments to enforcing gratuitouspromises or to third party claims, impediments which have since given wayin the United States. The obligation underpinning negligent misstatementresembles contract more than it does the tort of negligence (specifically itresembles contract theories of promissory estoppel and third party benefi-ciary) for a duty of care must be voluntarily undertaken, typically by anactor inviting a claimant to rely on information. I turn to this point nowfor it is the key to understanding the power of an actor to contract out ofliability for negligent misstatement when the misstatement is not aninducement to contract.

III. CONTRACTING OUT OF LIABILITY FOR NEGLIGENTMISSTATMENT

Recall that in Snyder v Lovercheck the buyer, Snyder, included a claimagainst his agent, Hayek, alleging that Hayek was negligent in failing towarn him that the exculpatory terms in the seller’s contract absolved theseller from responsibility for the accuracy of the representation that ryeinfected only 100 of the 1,900 acres purchased. The court held that Hayek

58 Ultramares Corp v Touche 174 NE 441 (NY 1931) and Glanzer v Shepard 135 NE 275(NY 1922) are the primary cases.

59 Hedley Byrne & Co v Heller & Partners Ltd [1964] AC 465 (HL). [Hedley Byrne]

260 Mark P Gergen



was not subject to negligence liability because Snyder could be expected tohave read and understood the implications of the exculpatory terms in theseller’s contract. It may seem odd that exculpatory terms in a contractbetween Snyder and the seller could absolve Hayek from negligenceliability to Snyder. This part explains why the result is correct. Hayek owedSnyder no duty of care because Hayek never appeared to invite Snyder torely on him for advice regarding the contract. That Snyder was abusinessman who could read and understand the exculpatory termssupports this conclusion.

Generally, when negligence may result in solely pecuniary harm, an actorhas a duty of care in supplying information to a recipient if and only if theactor invites the recipient to rely on the information.60 More precisely,there is a duty of care if the actor reasonably appears to intend to invite therecipient to attach significant importance to the information in making adecision that may result in a loss if the information is inaccurate ormisleading. I will call this ‘the requirement of invited reliance’. StephenPerry has called this ‘the requirement of an undertaking’.

That a duty generally requires invited reliance distinguishes the tort ofnegligent misstatement from the general tort of negligence. It is sufficientfor a duty of care in the general tort of negligence that an actor’s conductcreates a risk of harm to another that the actor could reduce by theexercise of reasonable care. When people drive on a public road they havea duty of care for the simple reason that driving creates a risk of harm toothers that a driver can reduce by driving with care. The existence of aduty does not depend on a driver appearing to intend to invite others torely on the driver being careful. On the other hand, when an actor givesinformation to a claimant bearing on a prospective business transactionbetween the claimant and a third person, the actor undertakes a duty ofcare to ensure the accuracy of the information only if the actor reasonablyappears to intend to invite the claimant to rely on the information.61

60 I make the case for this criterion of duty as an accurate description of United States lawin MP Gergen, ‘The Ambit of Negligence Liability for Pure Economic Loss’ (2006) 48Arizona Law Review 749, 755–60. A similar point has been made regarding Commonwealthlaw in SR Perry, ‘Protected Interests and Undertakings in the Law of Negligence’ (1992) 42University of Toronto Law Journal 247 and B Feldthusen, Economic Negligence, 4th edn(Scarborough, Carswell, 2000) 48–54, 120–21.

61 The case of Stewart Title of Idaho v Nampa Land Title Co Inc 715 P 2d 1000 (Idaho1986) illustrates this important distinction. An agent of a title company mistakenly told anagent of an escrow company over the phone that defects in title disclosed in a preliminaryreport had been cleared. The escrow agent released funds on this basis. The court found noliability because of a trade custom of giving written clearance. The defendant’s agent couldhave foreseen that the claimant’s agent might rely on the information given over the phone,but given the trade custom she did not reasonably appear to intend to invite the escrow agentto release funds on the basis of the phone call.




It follows from the requirement of invited reliance that informationsuppliers have the power to determine whether they have a duty of care,and to determine the content and scope of any duty they undertake, bysignaling to a recipient whether and how they intend the recipient to beable to rely on the information. Again this is in sharp contrast with thegeneral tort of negligence. A driver on a public road cannot avoid or limitthe duty of care by signaling to others that they should not rely on thedriver being careful.62

An information supplier can avoid undertaking a duty of care bywarning the recipient not to rely on the information.63 A contract thatdefines the scope of an actor’s engagement will absolve the actor fromnegligence liability for failing to advise a client about matters outside thescope of the engagement even though it might seem reasonable for theactor to speak up. For example, Nowell v Dawn-Leavitt Agency 64 holdsthat an insurance agent employed to obtain a standard homeowner’s policyhad no duty to advise a client of the need for flood insurance though theneed should have been apparent to her but perhaps would not have been tothe client.65 The holding in Snyder v Lovercheck that Hayek had no dutyto advise Snyder regarding the contract is analogous. A contract thatdefines the scope of an actor’s engagement may even absolve an actor fromnegligence liability for affirmatively supplying misleading information

62 Certain categories of actors are held to a lower duty of care because people who comeinto contact with them should know they cannot be expected to exercise ordinary care. Forexample, infants and minors are sometimes held to an age-specific duty of care for thisreason.

63 Kuehn v Stanley 91 P 3d 346 (Ariz App 2004) (purchaser of property did not justifiablyrely on ‘short form appraisal’ that stated it was intended only for use in financing). Cf FirstNational Bank of Newton Cty v Sparkmon 442 SE 2d 804 (1994) (duty ‘may be limited byappropriate disclaimers which would alert those not in privity with the supplier of informa-tion that they may rely upon it only at their peril’). Often a disclaimer is given effect byholding reliance in the face of a disclaimer to be unjustifiable. See Quinn v McGraw-HillCompanies Inc 168 F 3d 331 (7th Cir 1999) (Illinois law) (disclaimer accompanying bondrating that rating was ‘not a recommendation to buy, sell, or hold any such Bonds and may besubject to revision or withdrawal at any time’ makes claimant’s reliance unjustifiable); DakotaBank v Eiesland, 645 NW 2d 177 (Minn App 2002) (disclaimer that financial statementswere an unaudited compilation makes claimant’s reliance unjustifiable).

64 617 P 2d 1164 (Ariz App 1980).65 Carleton v Tortosa 17 Cal Rptr 2d 734 (Ct App 1993) (holding that real estate broker

who did not undertake to supply tax advice is not subject to negligence liability for failing toadvise a client that transactions exchanging property could easily be restructured to avoidtax). Murphy v Kuhn 682 NE 2d 972 (NY 1997) is a representative case holding that aninsurance agent has no duty to advise a client regarding a need for coverage when the agentdoes not undertake to render such advice. California imposes a duty on an agent to advise aclient regarding insurance only when the agent misstates coverage, the client requests specificcoverage, or the agent holds themselves out as having expertise in the specific field:Fitzpatrick v Hayes 67 Cal Rptr 2d 445 (App 1997). For additional cases see the Annotation88 ALR 4th 289.

262 Mark P Gergen



when the information concerns matters outside the scope of the engage-ment. For example, Hill v Bache Halsey Stuart Shields holds that astockbroker who gave bad investment advice was absolved of liability by acontract that limited the broker’s duties to carrying out the client’sorders.66 Auditors routinely define the extent of their duty to ensure theaccuracy of a financial statement by the character of their engagement. Forexample, First National Bank of Bluefield v Crawford holds that inundertaking to supply a ‘review report’ of financial statements an auditordoes not undertake a duty to verify the accuracy of the financial informa-tion supplied to it.67

Courts do not automatically enforce contract terms that try to precludeor limit an actor’s duty in supplying information. Usually when a contractterm is disregarded this is consistent with a rule that protects invitedreliance because a claimant was reasonable in believing an actor intendedto invite the claimant’s reliance notwithstanding the exculpatory term.Ryan v Kanne illustrates this.68 A prospective buyer of a small companywith poorly maintained financial records told an accountant employed toprepare the company’s financial statements that the accounts payable wereof particular concern to him and to use ‘every conceivable means todetermine the accounts payable’. The accountant orally assured the buyerthe accounts payable would be accurate within $5,000. Nevertheless, thefinancial statements delivered by the accountant contained prominentdisclaimers that they were ‘unaudited statements’ and that the accountantexpressed no opinion on the validity of the financial information reported.These disclaimers were belied by a note in the comments describing specificmeasures taken to check the validity of accounts payable. It turned out thatthe accounts payable were grossly overstated and that a competent auditwould have discovered this. The court found the accountant undertook aduty to use reasonable care to verify the accuracy of the accounts payablenotwithstanding the disclaimers because of his repeated specific assurancesto the contrary.

Less clear is when and how an actor may absolve himself of negligenceliability through an exculpatory term if the actor invites the claimant torely on information supplied by the actor. Hedley Byrne raises the issue.69

The claimant asked its bank to make inquiries to determine if a client couldbe relied upon to pay for advertising which the claimant was placing onbehalf of the client. The bank knew the claimant was on the hook to payfor the advertising if the client did not. The bank replied by a letter stating‘without responsibility on the part of the bank or its officials’ that the

66 790 F 2d 817 (10th Cir 1986).67 386 SE 2d 310 (W Va 1989).68 170 NW 2d 395 (Iowa 1969).69 Above n 59.




client was ‘good for its ordinary business engagements’. Hedley Byrne isimportant because it is the first English case recognising the possibility ofnegligence liability for misstatement. However, the court held that theexculpatory language absolved the bank of a duty. Later English casessupply an alternative rationale—while the bank undertook a duty of carein answering the inquiry the exculpatory term absolved the bank fromliability for breach of the duty.70 If the criterion of invited reliance defineswhen there is a duty of care in supplying information, then the alternativerationale is preferable in Hedley Byrne because the bank probably didappear to intend to invite the claimant to rely on its response in decidingwhether to continue to place advertising for the client.71

Typically when an exculpatory term is enforced the circumstances aresimilar to those in Hedley Byrne. The exculpatory term is clear andreasonable in the circumstances because the actor supplied the informationfor a fee that is small in relation to the liability risk. For example, while thecases are dated it seems fairly well established that an exculpatory term inDun & Bradstreet’s subscription agreement absolves the rating agencyfrom negligence liability.72

An exculpatory term is not enforced if the term is unclear or if absolvingthe actor from negligence liability would be unreasonable in the circum-stances. Estey v MacKenzie Engineering Inc illustrates this point.73 A homebuyer hired an engineering firm to inspect a house prior to purchase to

70 Smith v Bush [1990] 1 AC 831 (HL); Harris v Wyre Forest District Council [1989] 2 AllER 514 (HL).

71 United States cases recognise that a bank has a duty of care in supplying a gratuitouscredit reference if the bank has reason to know the claimant is considering whether to extendcredit to the subject. Berkline Corp v Bank of Mississippi 453 So 2d 699 (Miss 1984) is aleading case. See also MSA Tubular Products Inc v First Bank and Trust Co, Yale, Oklahoma869 F 2d 1422 (10th Cir 1989) (Oklahoma law); Nevada National Bank v Gold Star MeatCo 514 P 2d 651 (Nev 1973); TC Russler and SH Epstein, ‘Disclosure of CustomerInformation to Third Parties: When is the Bank Liable?’ (1994) 111 Banking Law Journal258, 270.

72 See Fidelity Leasing Corp v Dun & Bradstreet Inc 494 F Supp 786 (ED Pa 1980)(Pennsylvania law); Hong Kong Export Credit Insurance Corp v Dun & Bradstreet 414 FSupp 153 (SDNY 1975) (New York law); Xiques v Bradstreet Co 24 NYS 48 (Sup Ct 1893),affirmed 36 NE 740 (NY 1894). Duncan v Dun 8 F Cas 9 (CCED Pa 1879) holds the clausealso bars a claim for gross negligence. Globe Home Improvement Co v Perth AmboyChamber of Commerce Credit Rating Bureau 182 A 641 (NJ App 1936) also gives effect to ageneral disclaimer when information is supplied for a nominal payment. The case holds thereis no liability beyond recovery of the amount paid for negligence in supplying a credit reportwhere the report stated the information ‘is based upon information obtained in good faith bythe agent from sources deemed reliable, the accuracy of which, however, is in no mannerguaranteed.’ The opinion emphasises the small amount paid for the report. Gale v Value Line640 F Supp 967 (DRI 1986) holds that the statement ‘[factual] material is obtained fromsources believed to be reliable but cannot be guaranteed’ in a publication ranking securities isinsufficient to preclude liability for negligence while finding no duty because the informationwas misleading by omission.

73 927 P 2d 86 (Or 1996). Schaffer v Property Evaluations Inc 854 SW 2d 493 (Mo App1993) is similar.

264 Mark P Gergen



identify major structural defects and settling. The contract limited thefirm’s liability to $200, which was the price paid for the inspection. Thebuyer ended up incurring a substantial loss because of undiscoveredstructural defects that a competent inspection would have revealed. Thecourt held the exculpatory term did not shield the firm from liability inexcess of $200 on a negligence claim. It declined to establish a per se ruleand instead based its decision on the facts that the claimant was aconsumer, the exculpatory term did not explicitly cover a negligence claim,the magnitude of the fee paid for the inspection, and the importance of theinspection to the claimant. There also is some authority for a rule that anexculpatory term will not shield an actor from liability for gross negli-gence.74

The lesson to be drawn is that an actor generally has no duty of care indisseminating information that may affect a recipient’s dealings with thirdpersons. A duty arises only when an actor invites a recipient to rely on theinformation. Actors can determine by contract whether they have a duty,and its content and scope, by making apparent when, and to what extent,they invite reliance. But once an actor undertakes a duty it is not so easilydisclaimed. The pattern is a familiar one in contract law. People cangenerally determine when they are under a contractual obligation byavoiding an apparent commitment, but once they make an apparentcommitment they may absolve themselves of the attendant obligation onlywhen this is reasonable in the circumstances.

IV. CONCLUSION

The law of contract has a fairly well-defined relationship to the law ofdeceit because the practice of deceit is inimical to the practice of contract.The possibility of deceit undermines trust, which is essential to contract.Deceit and contract are so adverse that we are deeply sceptical as towhether a person ever would intelligently agree to bear the risk of beingdeceived about a contract. The best argument for enforcing exculpatoryagreements to preclude fraud claims is that the agreements shield honestpeople from unfounded claims of fraud. At bottom, the argument forenforcing exculpatory agreements grounds on a deeply pessimistic view ofhuman litigiousness, and of the capacity of courts to distinguish goodfraud claims from bad, and on a very optimistic view of private ordering.Even if one is sympathetic to these views, one should consider thepossibility that rules requiring that fraud be pled with specificity, and

74 This exception is stated in the cases cited above n 72 enforcing the exculpatory term inDun & Bradstreet’s subscription agreement.




proven by clear and convincing evidence, provide sufficient protectionagainst unfounded claims of fraud while at the same time enabling courtsto redress proven fraud.

The law of contract has a less well-defined relationship to bodies of lawoutside of contract that provide redress for inadvertent misrepresentation.This is largely attributable to the obscurity of the law of equitablerescission and restitution and the immaturity of the law of negligentmisstatement. I have argued that principles of contract law generallyshould control on these issues. Contract law rules that protect againstunfounded claims of extra-contractual representations—this is the work ofthe parol evidence rule in the United States—should also protect againstclaims in equity, restitution, or tort. And contract terms that allocate therisk of the inaccuracy of a representation to a claimant should also applyto claims in equity, restitution, and tort. Often in a negligent misstatementcase an actor’s obligation to a claimant is not based on a contract becausethe actor provides information to a claimant gratuitously or pursuant to acontract with another. In such cases the requirement for a duty of care ofinvited reliance generally enables an actor to determine the existence anddefine the scope of the duty to a claimant.

266 Mark P Gergen



11

Assignments, Trusts, Property andObligations

ANDREW TETTENBORN

I. INTRODUCTION

DESPITE THE FACT that two sizeable books have recently beenwritten about the assignment of contractual rights,1 the topic isnot something that overly excites contracts scholars. It is neverthe-

less a very important topic in practice,2 and also one that raises someawkward questions about the nature of the rights arising under anassignment, not to mention issues about the law of trusts and the disputedterritory lying between property and obligation.

Two English cases have recently thrown some of these points into sharprelief, and for that reason seem worth at least a brief article. The cases bothinvolve what looks like a narrow question, namely, how far (if at all) athird party may use processes similar, but not identical, to assignment inorder to enforce for his or her own benefit a contractual right otherwisenon-assignable.

In the first case, Don King Productions Inc v Warren,3 two boxingpromoters with substantial pugilistic portfolios went into partnership.They agreed at the time, and confirmed later, that the benefit of all theircontracts with individual boxers should be joint assets, held on trust for

1 M Smith, Law of Assignment: The Creation and Transfer of Choses in Action (Oxford,Oxford University Press, 2007) and G Tolhurst, The Assignment of Contractual Rights(Oxford, Hart Publishing, 2006).

2 In England and Canada, one need only mention in this connection the vital roleassignment plays in lending against receivables, in securitisation and in the working out ofmultiparty construction contracts. In the American context, one can add the extensive use ofassignment, especially of malpractice claims, in settling litigation, on which see, eg, R Walters,‘The Unwitting Attorney, the Desperate Client, and the Perpetuation of the New York PowerPlay: a Proposal to Ban Voluntary Assignments of Legal Malpractice Claims via New YorkGeneral Obligations Law Section 12–101’ (2005) 3 Cardozo Public Law, Policy and EthicsJournal 543.

3 [2000] Ch 291 (CA), aff’g [2000] Ch 295 [Don King].



the partnership. The arrangement, like many similar ones, subsequentlysoured, and the partnership fell to be dissolved. At this point the issuearose as to whether the rights under the contracts did indeed form part ofits assets. The partner who had brought in the more valuable contractsraised an ingenious argument that they did not. The agreements, hepointed out, depended fundamentally on the skill and flair of the promoterappointed under them, and were for that reason in their nature toopersonal to be assigned.4 For good measure most of them also containedexpress anti-assignment clauses, which were indisputably valid.5 It mustfollow, he argued, that the right to enforce them could not be vested inanyone other than the original contractor.

Lightman J disagreed with these contentions. One of his reasons wasrelatively uncontroversial, namely that however unassignable an obligationmight be in principle, there was no difficulty in enforcing a contractualagreement between partners to treat the agreement as if it was partnershipproperty for the purpose of deciding who got what when the arrangementcollapsed. This seems plainly correct, and will not be discussed further.Lightman J’s second reason, by contrast, is much more difficult. Even if acontract was of such a personal nature as to be unassignable, or containeda clear contractual prohibition on assignment, according to Lightman J thisdid not mean that there was any bar as a matter of law to the promiseeholding the benefit of the contract on trust for a third party. It might betrue that allowing such a trust de facto emasculated the principle thatpersonal agreements could not be assigned and gave potential quasi-assignees an end-run around otherwise effective anti-assignment clauses,since a trust of an obligation which bound the obligee to enforce it as thebeneficiary’s catspaw has a result to all intents and purposes indistinguish-able from that of an equitable assignment to that same beneficiary.Nevertheless, this was not objectionable as such, at least in the absence ofa specific provision forbidding not only assignment but the creation oftrusts. On appeal, the Court of Appeal upheld Lightman J’s reasoning,though Morritt LJ added, somewhat enigmatically, that although anotherwise inalienable obligation might be held in trust, this would not assuch give the beneficiary any right to interfere in its enforcement.6

4 Under the principle in Nokes v Doncaster Amalgamated Collieries Ltd [1940] AC 1014(HL).

5 Any lingering doubts on this were dispelled by the House of Lords in Linden GardensTrust Ltd v Lenesta Sludge Disposal Ltd [1994] 1 AC 85 [Linden Gardens]. The law inCanada is similar: Brio Beverages (BC) Inc v Koala Beverages Ltd [1999] 6 WWR 219(BCCA).

6 See Don King, above n 3, at 335–6.

268 Andrew Tettenborn



Don King was followed, in a relatively uncontroversial way, in twosubsequent English decisions.7 But neither of these directly involved thequestion of allowing trust beneficiaries to enforce unassignable contracts.This question did arise bluntly, however, when the Court of Appeal wasfaced with Barbados Trust Co v Bank of Zambia.8 Briefly, Bank ofAmerica (BOA) held Zambian government bonds, expressed by their termsto be assignable to banks but not to other financial institutions. BOAnevertheless purported to assign these bonds to a non-bank, and followinga series of further transfers they ultimately ended up in the hands ofBarbados Trust Co (BTC), another non-bank, which wanted to sue onthem. The restrictions on assignment in the bonds obviously amounted to amajor roadblock, meaning that the right to sue remained all along in BOA,but, prompted by BTC, BOA sought to repair this problem by declaringitself trustee of the bonds for BTC. BTC then sued the Bank of Zambiawearing its hat as trust beneficiary instead of assignee, following thestandard practice of joining BOA, the trustee, as nominal defendant.9 Bankof Zambia for its part argued that the declaration of trust was a meresubterfuge to circumvent the contractual limitations on assignment, andthat BTC should remain unable to obtain indirectly what it could not getdirectly. BTC did indeed fail to recover, on the ground that BOA’s own titlehad been defective (for reasons not relevant here) and that hence BTC’s fellwith it. However, a majority of the Court of Appeal also decided, reversingLangley J and explicitly endorsing the reasoning in Don King, that the baron assignment to non-banks was an irrelevance.10 A declaration of trust,the court held, was not the same thing as an assignment: the prohibition onassignment did not in its terms prohibit a declaration of trust, andwhatever the position might have been had it purported to do so, thatmeant that BTC’s title to sue as trust beneficiary was beyond attack. Incontrast, Hooper LJ dissented, making the obvious contrary point. What-ever analytical differences might exist between assignments and declara-tions of trust, he observed, the result of allowing BTC to sue as thebeneficiary of a trust of a debt was precisely the same as allowing it to sue

7 See Swift v Dairywise Farms Ltd [2000] 1 All ER 320 (CA), where a trust was pressedinto service to allow hypothecation to a lender of proceeds of an otherwise nontransferablemilk quota arising under European Union agricultural marketing legislation; John Taylors (AFirm) v Masons (A Firm) [2001] EWCA Civ 2106, which applied the rule in Keech vSandford (1726) Sel Cas T King 61, 25 ER 223 (Ch) to renewals of unassignable licenses.

8 [2007] EWCA Civ 148, [2007] 1 Lloyd’s Rep 495 (CA) [Barbados Trust].9 The practice was standard in that it was implicitly approved in Les Affréteurs Réunis

SA v Leopold Walford (London) Ltd [1919] AC 801 (HL). See also Performing Right SocietyLtd v London Theatre of Varieties Ltd [1924] AC 1 (HL); Vandepitte v Preferred AccidentInsurance Corporation of New York [1933] AC 70 (PC) 74 [Vandepitte].

10 Waller and Rix LJJ.

Assignments, Trusts, Property and Obligations 269



as the equitable assignee of that debt, and it was a curious rule of law that,having driven this remedy out of the front door, promptly let it in by theback door.

In this article I argue that Don King and Barbados Trust not only reacha woefully unrealistic commercial result but more importantly misunder-stand the nature of equitable assignment. In particular, the supposedcontrast between trusts and equitable assignments, I suggest, is a false one.

II. EQUITABLE ASSIGNMENT: THE PECULIAR ENGLISH APPROACH11

If you ask a civil lawyer, or for that matter a Scottish jurist, what anassignment is, the answer is relatively simple: it is a wholesale substitutionof creditors.12 A creditor C with a right to exact performance from adebtor D has voluntarily transferred that right to an assignee A, who isthereby substituted for C and able to pursue D.

Unfortunately the common lawyers, for reasons never adequatelyexplained, could not reconcile themselves to this idea that an obligationcontracted in favour of C could morph into one owed to A, even if C washappy for it to do so.13 Whatever C might intend, in that respect he or sheremained the creditor and the only person entitled to claim performance. Thiswas why, given the commercial necessity of allowing assignment under someguise or other, equity had to be roped in to fill the gap. Unfortunately, it did soby using two, or possibly three, disjointed approaches simultaneously.14

To begin with, equity recognised what today we call assignment proper.If C entered into some agreement with A envisaging the transfer from C toA of the benefit of C’s rights against D, equity intervened from theseventeenth century to ensure that the right was enforced,15 and that when

11 We are not, for obvious reasons, concerned here with statutory assignments under theprovisions of s 136 of the Law of Property Act 1925 (UK) 15 & 16 Geo V c 20 or itsCanadian analogues (eg Law and Equity Act RSBC 1996 c 253, s 36).

12 On French law see the bald statement in B Starck et al, Droit Civil: les Obligations, 6thedn (Paris, Litec, 1998) vol 2 § 1591 (cessionaire de ces créances ‘pourra s’en prévaloir’). Seealso J Ghestin, Traité de Droit Civil: La Formation du Contrat, 3rd edn (Paris, LibrairieGénérale de Droit et de Jurisprudence, 1993) § 733. For Scotland, see D Walker, Principles ofScottish Private Law, 2nd edn (Oxford, Clarendon Press, 1975) vol 2, 1750 (the effect ofassignation is ‘to divest the cedent completely and to put the assignee in his place, entitled tosue for the enforcement of the assigned right, … to receive payment and grant a gooddischarge therefore’). The third edition of the same book, perhaps oddly, seems to omit thislapidary description.

13 On which see W Cook, ‘The Alienability of Choses in Action’ (1916) 29 Harvard LawReview 816; W Holdsworth, History of English Law (London, Methuen & Co Ltd, 1925) vol7, 520–25; Smith, above n 1, at §§ 5.04–5.09. The chief grounds for complaint seem to havebeen the inherently personal nature of personal actions and a loosely-articulated fear of thebuying up of proceedings by the unscrupulous and litigious.

14 What follows is neatly explained in Smith, above n 1, at §§ 6.02 ff.15 Holdsworth, above n 13, at 519.




it was, A got the benefit of it. This was accomplished by allowing A to suein equity to force C to invoke his or her claim, and then adding a congeriesof ancillary equitable rights to protect A’s position in other ways.16 Afterfusion,17 A sued and joined C in a single proceeding, with the ancillaryrights directly protected.18

Secondly, there was a parallel development, though one with much thesame long-term result. Lawyers had long talked about a debt or contrac-tual right being an item of property:19 in other words, an asset, and hencesomething that might figure in the ‘things’ section of a putative civil codeas well as under the ‘obligations’ rubric. As a result, there had logically tobe a role for the law of trusts proper. If a piece of land could be held ontrust, with legal property in one person and beneficial ownership inanother, the same must follow for a debt or other contractual right: itcould be owned in equity by someone other than the person appearing tobe the obligee.20 It followed that a contractual right could also be assignedde facto by the simple mode of C declaring himself or herself a bare trusteeof it for A.21

We are primarily concerned with assignments and declarations of trust.However, it is worth adding for the sake of completeness a third possibleform of transfer arising from the incorporation into the law of assignmentof the maxim that equity regards as done that which ought to be done. Itbecame clear by the end of the nineteenth century that a contract for valueby C to transfer a right—whether present or future—to A caused the rightin question to stand transferred in equity either immediately or as andwhen it arose.22 This rapidly became the most significant form of assign-ment, as witnessed in the growth of debt factoring and general chargesover book-debts, both of which depend on it. But whether a promise toassign is regarded as equivalent to an assignment proper of the right, or a

16 For example, by enjoining D from pleading a set-off incurred, or a release arranged withC, after D knew of A’s position as assignee. See R Derham, Set-off, 3rd edn (Oxford, OxfordUniversity Press, 2003) §§ 17.02–17.03.

17 Or rather, in the English context, from the time of the partial fusion effected by theCommon Law Procedure Act 1854 (UK) 17 & 18 Vict c 125, allowing common law courts togive effect to equitable interests.

18 For example, if D had a set-off against C incurred after notice of the assignment, thiswould simply be ignored and judgment given for A.

19 See, eg, Fitzroy v Cave [1905] 2 KB 364 (CA) 372–3 (Cozens-Hardy LJ); Ellis vTorrington [1920] 1 KB 399 (CA) 411 (Scrutton LJ); Tolhurst, above n 1, at § 3.20.

20 For a straightforward expression of this idea see, eg, Fletcher v Fletcher (1844) 4 Hare67, 67 ER 564 (ChD) 74 (Wigram V-C).

21 Smith, above n 1, at §§ 6.33 ff. It is also possible, of course, for a trust of a contractualright to arise out of some more complex trust relation, where the trustee has active duties. Astraightforward example is a trustee for bondholders. But this article is concerned with thetrust as a simple means of de facto assignment: ie a bare trust and no more.

22 As in the rule in Holroyd v Marshall (1862) 10 HL Cas 191, 11 ER 999. Its extensionto debts generally was cemented by Tailby v Official Receiver (1888) 13 AC 523 (HL).




bare trust of it (if there was a difference between the two) has never beenmade clear.23 One suspects that this is because the point does not oftenmatter in practice. In either case A is entitled in equity to have the rightexercised for his or her benefit and not C’s, even if, for example, C isinsolvent, and this is normally all that matters.

III. TRUSTS AND EQUITABLE ASSIGNMENTS: TWO LEGAL RESULTS ORONE?

Having dealt with the theory of equitable assignments, we can now turn toassignment and the creation of a bare trust in more detail. As describedabove, these are clearly two different processes. One simply requires anagreement, however expressed, to transfer the benefit of a contractual orother right to someone else.24 The other involves meeting all therequirements—certainty of intention, certainty of subject matter, and soon—necessary for the creation of a valid express trust,25 which is not thesame thing at all. But this is not a point that matters here. The importantquestion is not whether assignments and trusts are generated in differentways, but rather whether they also give rise to sufficiently different legalresults to justify the law’s discriminating between them, as Don King andBarbados Trust clearly suggest.26 It is my suggestion that, if we comparethem properly, they do not.

We begin by looking at the effects of an equitable assignment proper.These, it is suggested, can be effectively summed up in five propositions.First, A prevails, like any other valid transferee, against the creditors of C ifC is insolvent.27 Second, the best explanation as to why A prevails is thatfrom the moment of the equitable assignment C is treated by equity asholding the benefit of the obligation assigned on constructive trust for A as

23 But the latter is perhaps more likely. In particular, Lord Fitzgerald in Tailby v OfficialReceiver, ibid, at 546 based the rule on the idea that ‘whenever persons agree concerning anyparticular subject, that, in a Court of Equity, as against the party himself, and any claimingunder him, voluntarily or with notice, raises a trust’ (emphasis added).

24 See, eg, William Brandt’s Sons & Co v Dunlop Rubber Company Ltd [1905] AC 454(HL) 462 (Lord M’Naghten).

25 On which see, eg, J McGhee (ed), Snell’s Equity, 31st edn (London, Thomson, 2005) §§20–15.

26 An analogy may make this clearer. At common law a gift of a chattel may be made intwo entirely different ways: by delivering it or alternatively by executing a deed of gift.Nevertheless, the legal result of these two different actions is precisely the same: the doneebecomes owner of the chattel concerned and gets all the rights one would expect an owner tohave.

27 The assignee is, of course, subject to specific limitations, such as the insolvency rulesproviding for the annihilation of transfers in fraud of creditors. But so is the transferee of anyother asset, so this changes nothing.




beneficiary.28 Third, A can insist—originally by separate proceedings for acommon injunction, but since the fusion of law and equity by proceedingsdirectly against D,29 with C joined as a formal party30—that the right beexercised for his or her benefit, and hence take control of its enforcement.Fourth, A takes ‘subject to equities’. More precisely, this means (i) that Acan only enforce the right ‘warts and all’, that is, subject to any defences orlimitations31 and (ii) that A will be subject to certain rights of set-off—broadly, all connected counterclaims32 and certain unconnected onesarising before notice of the assignment.33 Fifth, A has the right toprotection from any post-notice set-off arising between assignor anddebtor34 and any post-notice agreement between assignor and debtor torelease or change the contractual rights involved.35

Now, how far do these legal results apply to a bare express trust of adebt? Let us take each in turn.

A. Prevalence in Insolvency

The interests of any trust beneficiary prevail against the creditors of aninsolvent trustee, and choses in action are no exception. We do not need totake this point any further.

28 Smith, above n 1, at § 6.12. But cf J Story, Commentaries on Equity Jurisprudence, 3rdEnglish edn (London, Sweet & Maxwell, 1920) 432, which regards an assignee as equivalentto the beneficiary of an express trust.

29 Or, more accurately in the English context, since the Common Law Procedure Act1854, above n 17, allowed common law courts to give direct effect to equitable rights.

30 Durham Brothers v Robertson [1898] 1 QB 765 (CA) 770 (Chitty LJ); PerformingRight Society Ltd v London Theatre of Varieties Ltd [1924] AC 1 (HL) 19 (Viscount Finlay).The fact that the original creditor must be joined in the action is a side-issue here, being onthe modern view largely a matter of procedure: eg The Aiolos [1983] 2 Lloyd’s Rep 25 (CA)32, 34; Weddell v Pearce & Major [1988] 1 Ch 26. See also n 9 above.

31 See, eg, Athenaeum Life Assurance Co v Pooley (1853) 3 De G & J 294, 44 ER 1281(CA); Graham v Johnson (1869) LR 8 Eq 36.

32 Newfoundland Government v Newfoundland Railway Co (1888) LR 13 AC 199 (PC).More precisely, the opposable claims are those that between D and C would amount toequitable or ‘transactional’ set-off.

33 Watson v Mid Wales Railway Co (1867) LR 2 CP 593. More precisely, this coversclaims otherwise covered by the principles contained in the Statutes of Set-off 1729 and 1735(2 Geo II c 22 and 8 Geo II c 24 respectively). The reason for limiting this susceptibility topre-notice cross-claims is that a debtor who gives credit to the creditor not knowing of anassignment should not lose any rights of set-off he or she would otherwise get, but a debtorwho knows of the assignment deserves no such indulgence.

34 Ibid.35 Brice v Bannister (1878) 3 QBD 569 (CA).




B. The Nature of the Beneficiary’s Interest in the Obligation Concerned

As mentioned above, the interest of an equitable assignee, on the bestexplanation, arises from a constructive trust.36 That of a trust beneficiaryproper arises, ex hypothesi, from an express trust. There is clearly adifference here. Nevertheless, it is a difference of source rather than result.In both cases the assignee or beneficiary ends up as the equitable owner ofthe obligation, and in both cases can enforce the obligation in the sameway—namely by the simple expedient of joining the assignor or trustee.

C. Control over Enforcement

Logically, this feature—the ability of the beneficiary to control and ifnecessary to compel enforcement—must apply in the trust scenario asmuch as in the case of assignment, at least as concerns a bare trust (it maybe different where the terms of a trust leave some discretion to the trusteesas to how to handle the trust assets: we will return to this later). Thiscertainly seems to follow from the cases on trusts of a promise generally,such as Vandepitte v Preferred Accident Insurance Corporation of NewYork.37 There the Privy Council implicitly approved38 the procedurewhereby, assuming the benefit of a contractual obligation is indeed held ontrust,39 proceedings can be brought by the beneficiary in his or her ownname, with the trustee joined as co-plaintiff or co-defendant as required.This procedure, moreover, was approved in Barbados Trust itself.40 More-over, this makes good sense. For a bare trustee of a contractual right torefuse point-blank to enforce it when asked to do so by the beneficiary41—and hence in effect to nullify it—would seem to be as blatant a breach oftrust as one could imagine. If so, it seems fairly obvious that the beneficiaryought to be able to prevent such a breach of trust by some device such asthat described. The choice, in other words, as to whether the obligation isenforced lies squarely with the plaintiff.

The only possible counter-argument comes in a slightly curious throw-away suggestion by Lightman J in Don King that the beneficiary under atrust might not in fact be able to control enforcement.42 This, he reasoned,was because the rule in Re Brockbank said that a beneficiary could not

36 See above, text accompanying n 28.37 Above n 9.38 Ibid, at 79 (Lord Wright).39 In fact it was not so held in Vandepitte, but that does not alter the point being made

here.40 See in particular Rix LJ in Barbados Trust, above n 8, at [99].41 Assuming adequate arrangements made for costs and the like.42 Above n 3, at 321.




dictate to a trustee how to exercise his functions.43 But this must bemisconceived, for two reasons. First, the point of Re Brockbank is not thata beneficiary cannot insist on a trustee observing the terms of the trust, butrather that a beneficiary, even if solely entitled, cannot tell the trustee tocommit an act that would amount to a breach of trust.44 If there is nodiscretion given to the trustee—and it is suggested that a bare trust of anobligation cannot leave it up to the trustee to decide whether to enforceit—there is no reason for the rule to apply.45 Second, if Lightman J is right,and the beneficiary of a trust of a chose in action cannot insist that thetrustee actually exercise it, this seems almost entirely to defeat the point ofthe trust in the first place. Put shortly, if the beneficiary does not have thisright, what rights worth having does he possess?46

D. ‘Subject to Equities’

What of the fourth principle, the ‘subject to equities’ rule? One aspect ofthis is easy: the rule that the benefit of an obligation can only be takenwarts and all must apply in the law of trusts. Although there appears to beno direct English authority on the application of this rule to trustbeneficiaries (as against assignees),47 it must be the case that the benefici-ary of a trust of a right gets the right to sue that the trustee has, no moreand no less. You cannot declare yourself trustee of something you do nothave. It follows that in so far as an obligor has an excuse for refusingperformance, such as an ability to plead non-liability due to fraud or error,or breach of fiduciary duty, this right must remain.48

What of the other aspect of the ‘subject to equities’ rule—counterclaimsavailable to the debtor? In so far as connected counterclaims are concerned(such as those arising out of equitable set-offs which would have been

43 [1948] Ch 206.44 In that case, the beneficiaries had purported to tell a surviving trustee whom to appoint

as a new trustee, something which was specifically left to the trustee’s own discretion.45 Cf Citibank NA v MBIA Assurance SA [2007] 1 All ER (Comm) 475, where the Court

of Appeal certainly assumed that a trustee may, if the terms of the trust so provide, follow thedirections of a beneficiary in exercising rights against third parties.

46 See Millett LJ in Armitage v Nurse [1998] Ch 241 (CA) 253: ‘there is an irreduciblecore of obligations owed by the trustees to the beneficiaries and enforceable by them which isfundamental to the concept of a trust. If the beneficiaries have no rights enforceable againstthe trustees there are no trusts.’

47 But there is authority concerning those entitled to the benefit of contractual rights bysubrogation: see The Front Comor [2005] 2 All ER (Comm) 240 (QB) and The Jay Bola[1997] 2 Lloyd’s Rep 279 (CA) holding that if a right can only be exercised by the originalcontractor subject to an arbitration clause, the same goes for an insurer claiming bysubrogation. And if a beneficiary takes subject to certain counterclaims (see Murphy vZamonex Pty Ltd (1993) 31 NSWLR 439 (SC)) then it must follow a fortiori that he takessubject to defences.

48 This also seems to be implicit in Murphy v Zamonex Pty Ltd, ibid.




available as between debtor and trustee), logic dictates that they must beopposable to a trust beneficiary as much as to an assignee. To begin with,in so far as they would have amounted to an equitable defence against thecreditor, the same reasoning applies as to the other defences mentionedabove. In any case, the rationale for allowing such a set-off is that theconnection between the right being sued on and the debtor’s cross-claim isso close that it would be inequitable to the debtor to determine the formerwithout the latter. If this is right, it should make no difference that theperson seeking to enforce the right is acting as trustee for someone elsewho actually stands to benefit. The injustice to the debtor is, after all,precisely the same in both cases. Indeed, this has been held to be the caseboth in Australia49 and in England.50

What about unconnected counterclaims against the trustee? In theassignment context, these are (as mentioned above) available against anassignee if created before the debtor knew of the assignment, on the basisthat they would have afforded a defence under the Statutes of Set-off hadthe assignor sued, and that an assignment of which the debtor wasunaware at the relevant time should not alter the position. There seems tobe no authority deciding whether the same thing applies to a trust of theobligation, but the reasoning in the assignment cases must apply heretoo.51 A debtor who thinks he or she is obtaining a set-off ought to be nomore affected by an unknown trust than by an unknown assignment.

E. Insulation from Post-Notice Equities

The question as to whether a beneficiary under a trust is insulated frompost-notice cross-claims or changes is a little more awkward. In the contextof the assignee, what is now the established rule arose in the days beforefusion from the Court of Chancery’s practice of supporting the assignee’sright with certain ancillary orders. These notably included injunctionsrestraining the debtor from relying, in any action by the creditor, on eitherthe Statutes of Set-off (in the case of a cross-claim otherwise covered by

49 Murphy v Zamonex Pty Ltd, ibid. The case involved a claim on a loan by a trusteesuing as such. The court held that a trade practices claim by the borrower against the trustee’spredecessor could be pleaded as equitable set-off. See also Doherty v Murphy [1996] 2 VR553 (SC).

50 Penwith District Council v VP Developments Ltd [2005] EWHC 259 (Ch) (Laddie J).An unsatisfied arbitration costs order had been made against a construction company. Thecompany then claimed in related arbitration to recover alleged underpayments. This claimwas available for set-off so as to prevent a winding-up on the basis of the costs order,irrespective of the fact that the company was insolvent and suing entirely as trustee for benefitof its creditors.

51 And indeed it is assumed to do so in Derham, above n 16, at § 17.96, where it is stated(though without authority) that the rules of statutory set-off are the same for obligations heldon trust as for assigned ones.




them) or, as the case might be, on any agreement to release or reduce therights of the original creditor.52 Since fusion, life is more straightforward:these rights of the assignee are directly recognised by the courts and theset-offs that would otherwise be available to the debtor are ignored. Butcan the same reasoning be applied to rights simply held on trust?

At first sight, it might indeed seem not, thus providing at least some clearblue water between trusts and assignments. Consider a common kind oftrust, such as a trust of a landed estate or a business, or a commercial trustsupporting an issue of bonds whereby one custodian holds the legalinterest with the actual investors as beneficiaries. No one would seriouslyargue that a debtor to such a trust, even if he knew he was dealing with atrustee, should be unable to deal with, and if necessary compromise, thetrust’s claims against him without the express permission of every benefi-ciary. If this were the law, it would negative the whole point of sucharrangements, which is to leave the day-to-day management of the businessor portfolio to the trustees’ informed discretion. Moreover, it can also bepointed out that in at least one arguably analogous situation, the benefici-ary is not protected. Whereas an assignment plus notice to the debtor locksthe assignee’s rights in against subsequent alterations, the process ofinsurer’s subrogation to claims of the assured does not. On the contrary, itis well-established that even where an insurer does have subrogation rightsover the assured’s claim against the obligor, those rights can be validlyreleased or altered by the assured, and if they are, the insurer’s only remedyis against the assured.53

Whether these points will hold water where we are dealing with a baretrust of an obligation, however, is highly doubtful.

As for the point about a trust of a business or a bondholders’ trust, theresponse is that not all trusts are the same. The functions of trustees, and withthem the duties they owe to beneficial owners, vary. In particular, there isevery difference between the kinds of trust mentioned in the previous para-graph and the kind of bare trust that does duty as a surrogate assignment.Complex trusts by their nature—and not infrequently by expressstipulation—give the trustee an active role in managing the trust property,and the discretion that goes with that role.54 A bare trust, by contrast, is justthat: rather like a well-trained footman, the trustee has little or no functionbesides holding the legal interest for the benefit of the cestui que trust. And ifthe trustee has no discretion or real independent function, then third partiesdealing with the trustee have no reason to expect him or her to be able torelease the beneficiary’s rights without the latter’s consent.

52 See above, n 16.53 West of England Fire Insurance Co v Isaacs [1897] 1 QB 226 (CA).54 Even on occasion to the extent of disregarding the express wishes of the beneficiary: see

Re Brockbank, above n 43.




Furthermore, it is suggested that the analogy between bare trusts and theinsurer’s right of subrogation is false. First, the assured retains a realinterest on his or her own behalf in the claim even where there issubrogation: there may, for example, be uninsured losses, or the possibilityof recovery over and above the amount received from the insurer. Insur-ance subrogation is therefore far removed from the kind of bare trust weare considering here, whose object is simply to transfer the benefit of theobligation lock, stock and barrel to the beneficiary. Second, it is also ratherdubious whether subrogation creates a true trust of the obligation infavour of the insurer in any case. Although the matter is not beyondargument, the better position seems to be that the insurer gets a proprietaryright over the proceeds of the right, but not over the right itself.55 If so, it isnot surprising that the right itself, remaining beneficially owned by theassured, can be altered by him or her at will.

Admittedly, in the absence of clear pre-fusion authority on what theCourt of Chancery would have done about a plea of set-off or release inthe case of a bare trust, it is difficult to be absolutely certain as to what thelaw is. Nevertheless, what little authority there is in England suggests thatwhere an obligation is, to the knowledge of a third party, held on a baretrust, then the beneficiary is indeed protected from alterations to his or herdetriment.56

IV. MORE ON TRUSTS AND ASSIGNMENTS

Pulling together the above strands, it is difficult to see any difference ofimportance between an assignment and a trust. The only distinction found,that one creates a constructive trust and the other an express trust, may betechnically genuine, but its rational significance is nil.57 If this is correct,then the major plank in the recent decisions on assignment—that there is

55 Re Ballast plc [2007] BCC 620 (Ch). See also Morris v Ford Motor Co Ltd [1973] QB792 (CA) 800, where Lord Denning MR doubted if subrogation was anything likeassignment.

56 See Jessel MR in Re Empress Engineering Co (1880) 16 ChD 125 (CA) 129: ‘A mereagreement between A. and B. that B. shall pay C. (an agreement to which C. is not a partyeither directly or indirectly) will not prevent A. and B. from coming to a new agreement thenext day releasing the old one. If C. were a cestui que trust it would have that effect’(emphasis added). See also Smith, above n 1, at § 13.11 and cf AW Scott, Scott on Trusts, 3rdedn (Boston, Little, Brown & Co, 1967) vol 4, 2517 to the same effect. There is oneapparently contrary authority, Gibson v Winter (1833) 5 B & Ad 96, 110 ER 728 (KB) butthis was doubted by Lord Campbell in De Pothonier v De Mattos (1858) El Bl & El 461, 120ER 581 (KB) 483 and by the Supreme Court of Canada in Culina v Giuliani [1972] SCR 343,22 DLR (3d) 210.

57 It is true that Waller LJ in Barbados Trust, above n 8, at [43] also observed anotherdifference, namely that an equitable assignment, unlike a trust, might be convertible into astatutory assignment under s 136 of the Law of Property Act 1925, above n 11, by the givingof a suitable written notice. This is correct. But can this matter?




some special feature of equitable assignment that makes it different fromother equitable interests—disappears.

Before we conclude that Don King and Barbados Trust were wrong toallow trust beneficiaries to subvert the rule against assignment of non-assignable rights, however, there is one more point that needs to beaddressed. Even if trusts and equitable assignments are effectively thesame, could it be that the real illogic lies not in allowing indirectenforcement by the beneficiary of a trust of an otherwise unassignableobligation, but in the rule allowing anti-assignment clauses to invalidateequitable assignments in the first place? However non-physical or legallyfictitious they may be, contractual rights are, after all, a kind of property.And since people can generally make what arrangements they like withtheir own property and alienate it as they think fit, it follows that weshould recognise the rights of the beneficiary of a trust of a right, even ifthat right is purportedly unassignable.58 Put another way, equity inrecognising an assignment does not transfer the right from creditor toassignee, but rather leaves it in the assignee while acting in personamagainst the creditor to force the latter to give effect to the (property) rightshe or she has purported to grant. And if this is right, then while anassignment in breach of an anti-assignment clause may be a breach of thecreditor’s contract with the debtor, there is no reason why it should affectthe assignee’s rights against the creditor.59

The answer to this, it is suggested, is threefold. First, although courtssometimes behave as if an equitable assignor indeed remains the owner of theassigned obligation,60 with the assignee’s rights lying against him or heralone,61 it is suggested that equity’s attitude to assignment went further thanthis and was more consistent with a view that the assignee should, as far aspossible, be regarded as having the creditor’s right actually transferred to him

58 Lightman J expressed this thinking perfectly in Don King, above n 3, at 321: ‘I can seeno objection to a party to contracts involving skill and confidence or containing non-assignment provisions from becoming trustee of the benefit of being the contracting party aswell as the benefit of the rights conferred.’

59 A position well-expressed, though ultimately rejected, in RM Goode, ‘InalienableRights?’ (1979) 42 MLR 553. See also its discountenancing in Linden Gardens, above n 5.There are also signs of it in Foamcrete Ltd v Thrust Engineering Ltd [2002] BCC 221 (CA),on which see A Tettenborn, ‘Prohibitions on Assignment–Again’ [2001] Lloyd’s Maritime andCommercial Law Quarterly 472.

60 See, eg, Warner Bros Records Inc v Rollgreen Ltd [1976] QB 430 (CA), holding that thepower to exercise an option under a contract remains exclusively in the original contractordespite equitable assignment. We ignore in this context the need to join the assignor in anyaction against the debtor, since this is today largely a formal requirement: see the authoritiesreferred to above, n 30.

61 Thus reproducing the classical view that equity does not subvert legal rights or titles,but rather recognises them and then puts constraints on their exercise: eg FW Maitland,Equity: A Course of Lectures, 2nd edn (Cambridge, Cambridge University Press, 1949) 106,149.




or her. In particular, equity in pre-fusion days was prepared to act, not onlyagainst the assignor, but against the debtor directly, for instance by enjoininghim or her from pleading set-offs arising after notice of assignment, or onoccasion ordering him or her to account in equity (that is, to pay) to theassignee notwithstanding prior payment to the creditor.62

Second, even against the background of the rule that one can declareoneself trustee of any asset whatever, the present position of equity asregards assignments of non-assignable rights—that it will procure that, asbetween assignor and assignee, any benefits received by the assignor arehanded to the assignee,63 but that it will not ensure that the right isenforced against the obligor64—is not as unprincipled as it might seem.The reason is that there is no equity in granting a remedy that will have theeffect of rendering nugatory a legitimate third party right,65 in this caseright of the debtor to decide with whom he or she is willing to enter intocontractual relations. Since, as has been argued above, there is no differ-ence worth the name between trusts and assignments, it seems to followthat exactly the same argument ought logically to apply to express trusts.Indeed, Don King and Barbados Trust themselves seem to go at leasthalfway towards this position. There are obiter dicta in both cases thatsuggest there may be at least some limits to the enforcement of an expresstrust of an obligation. In particular, both contemplate (but pointedly donot decide) that an express stipulation in a contract that no trust of it shallbe created or recognised may be given effect against the purportedbeneficiary of any such trust.66 If they are prepared to go that far, itremains to ask why they do not draw the logical conclusion and denyexpress trusts enforcement in so far as it would defeat the obviousexpectations of the parties.

Third, and more generally, despite the underlying practice of Englishlawyers in classifying choses in action and their transfer as part of the lawof property67 (and hence, by transference, as an ordinary subject of the law

62 Malcolm v Scott (1847) 6 Hare 570, 67 ER 1290 (ChD).63 Glegg v Bromley [1912] 3 KB 474 (CA).64 Linden Gardens, above n 5.65 For analogous situations, see, eg, equity’s clear power to refuse specific performance

where the effect of giving it would be to defeat a contractual stipulation entered into by thedefendant in favour of a third party (Warmington v Miller [1973] QB 877 (CA)) and to refuseto lend its aid to an equitable chargee where, to the knowledge of the chargee, this woulddefeat a negative pledge clause previously agreed with a third party (English & ScottishMercantile Investment Co Ltd v Brunton [1892] 2 QB 700 (CA)).

66 See Lightman J in Don King, above n 3, at 321; Rix LJ in Barbados Trust, above n 8, at[88].

67 As to this practice generally, and why it may have arisen, see Tolhurst, above n 1, at15–17, 55–6, 62. In particular, it has made it easier for the law to accommodate a wide ruleof prima facie transferability, with the attendant advantages to financing.




of trusts), there is also—as can be seen in the civil law approach68—a verydistinct whiff of the law of obligations here. This is especially true wherethe right being transferred arises from a promise or other voluntarytransaction on the part of somebody else.69 In such cases, the concern isnot simply with the relationship between a person and a thing: on thecontrary, the other side of the right—the obligor’s duties—are crucially inplay. To whom, it is vital to know, does the obligor owe performance?With whom must he or she negotiate to cancel or reduce exposure? Can heor she choose to whom he or she is to be bound? And if so, then the termsof the obligation involved become, not some side-issue to the law of trustsor some matter as between debtor and creditor alone, but the central issueof legal policy.

It is therefore respectfully suggested that if, as seems to be the case, thereis no difference of substance between equitable assignments and trusts ofan obligation, one immediate conclusion is justified: in so far as assign-ments are prohibited by the nature of the obligation concerned, or itsexpress terms, then there is no argument from logic in allowing that rule tobe avoided by using the device of a trust. Nor is there much of a practicalargument for it either. The justification for the English practice of allowingalmost unlimited rein to anti-assignment clauses is thoroughly commercial:a debtor should be able to choose to make sure that if he or she goes tosleep next to Portia, he or she will not wake up entwined in the arms ofShylock,70 whether by trust, assignment or anything else. The argument ofthe Court of Appeal in both Don King and Barbados Trust that (in effect)it was all the boxers’ or Bank of Zambia’s fault for not prohibiting bothassignments and declarations of trust in so many words is on this countremarkably thin. Contracts and similar arrangements are, after all, to beinterpreted in a reasonable, business-friendly and non-technical way, as theHouse of Lords has said71 and Lightman J in Don King indeed accepts.72 Itis difficult, with respect, to think of a less businesslike, literal andnit-picking interpretation than one which attributes to contractors, as DonKing and Barbados Trust do, an intention that a party to a contract shouldbe protected from proceedings by a third party wearing the hat of an

68 Tellingly, cession comes under the law of sale (effectively the law of obligations) in theCode Civil: see §§ 1689–1691. The relevant provisions in the Québec Civil Code (§§ 1637 ff)also appear under the law of obligations.

69 I put it this way, rather than referring to the more natural ‘contractual rights’, becausethe same argument applies in respect of some obligations, such as those of professionalcompetence, that can as a result of Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 beclassified as either contractual or tortious.

70 See Linden Gardens, above n 5, and in Canada Brio Beverages (BC) Inc v KoalaBeverages Ltd, above n 5.

71 Investors’ Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 AllER 98 (HL) 114–15 (Lord Hoffmann).

72 See Don King, above n 3, at 310–11.




assignee but not from proceedings by that same third party in the guise ofa trust beneficiary. The House of Lords, or any other Commonwealthsupreme court faced with this problem, as surely they will be, would dowell to bear this in mind.




12

The Nature of Equitable Assignmentand Anti-Assignment Clauses

C H THAM*

I. INTRODUCTION

ANTI-ASSIGNMENT CLAUSES are prevalent in modern com-merce. This is due, perhaps, to the reluctance of contracting partiesto deal with third parties with whom they may have had no prior

relationship. The key English decision giving voice to this concern is that ofthe House of Lords in the appeals heard in Linden Gardens Trust Ltd vLenesta Sludge Disposals Ltd1 where it unanimously2 took the view thatthe anti-assignment clause common to both appeals before it invalidatedthe purported equitable assignments3 for value of the benefits of thebuilding contracts in question. But in accepting that the assignments in thetwo appeals before them were ineffective, their Lordships opened up a‘black hole’ of an incompensable loss,4 since the would-be assignees werethereby merely third parties to the building contracts which the contractors

* Thanks are owed to Adrian Briggs, without whose encouragement this articlewould not have been written. All errors, however, remain mine alone.

1 [1994] 1 AC 85 (HL) [Linden Gardens], being the conjoined appeals of LindenGardens Trust Ltd v Lenesta Sludge Disposals Ltd (1990) 25 Con LR 28 (QB), rev’d (1992)30 Con LR 1 (CA) and St Martins Property Corporation Ltd v Sir Robert McAlpine Ltd(1991) 25 Con LR 51 (QB), rev’d (1992) 30 Con LR 1 (CA) [St Martins].

2 Lord Griffith’s divergence was only on the issue of the manner in which the employerin one of the appeals (St Martins) could recover substantial damages from the buildingcontractors.

3 Though none of the speeches make this explicit, it is fairly clear that their Lordshipswere concerned with the effectiveness of the anti-assignment clauses vis-à-vis equitableassignments because in one case (Linden Gardens) notice of the assignments was never givenand in the other case (St Martins) notice was only given long after the specific time underconsideration. Since statutory assignment only occurs on receipt of written notice of theassignment, by implication both appeals must have been decided by reference to the equitableassignment of the relevant legal choses in action.

4 It is probably true that this ‘black hole’ would still have manifested itself had therebeen no anti-assignment clause, or had effect not been given to the anti-assignment clauses:see Darlington Borough Council v Wiltshier Northern Ltd [1995] 1 WLR 68 (CA). But had



had breached. Having created the black hole, though, the majority inLinden Gardens promptly filled it back in by relying upon an extension ofa common law exception developed by Lord Diplock in The Albazero.5

Leaving the science of black holes for another day, this article aims tore-examine the nature of assignments, in particular equitable assignments.It asks whether the general acceptance of anti-assignment clauses aseffective to invalidate equitable assignments might not be due for anoverhaul. The proposition in this article is that once we clarify what anequitable assignment of a contractual chose in action entails, the under-standing of, inter alia, the proper role of an anti-assignment clause maywell be ripe for change.6

II. EQUITABLE ASSIGNMENT OF BENEFITS

In England, that an anti-assignment clause may invalidate an otherwiseeffective assignment between promisee-assignor and third party-assignee isseemingly well accepted, following the decision of the House of Lords inLinden Gardens. There, the building contracts between the purportedassignors and the obligors were on the Joint Contracts Tribunal (‘JCT’)standard form, and clause 17(1) prohibited assignments without the priorwritten consent of the obligor.7 In Lord Browne-Wilkinson’s view, ‘clause17(1) of the [JCT standard form] contract prohibited the assignment by theemployer of the benefit of the contract. This, by itself, is fatal to the claimby [the assignee] in the St Martins case.’8 Why? In Lord Browne-Wilkinson’s view, (purported) assignments without the consent of thebuilding contractors constituted a breach of clause 17,9 and such assign-ments were ineffective to vest any causes of action in an assignee.10

Lord Browne-Wilkinson seems to have taken Tom Shaw and Co v MossEmpires Ltd11 as standing for the proposition that a prohibition onassignment could invalidate the assignment as against the other party tothe contract so as to prevent a transfer of the chose in action to a third

the House of Lords upheld the validity of the assignments in both appeals in Linden Gardens,the problem of the black hole would have become entirely academic.

5 [1977] AC 774 (HL). Lord Griffith arrived at the same conclusion but through adifferent route.

6 Assuming, of course, that the equitable assignment is not otherwise barred as beingvoid for maintenance or champerty.

7 The parties in St Martins were dealing with each other on the basis of the 1963 edn(July 1972 rev) form, whereas the parties in Linden Gardens were dealing with each other onthe basis of the 1963 edn (July 1975 rev) form. Both versions of the JCT standard form hadan identically worded clause 17(1).

8 Linden Gardens, above n 1, at 103.9 Ibid, at 106.

10 Ibid, at 107–108.11 (1908) 25 TLR 190 (KB).

284 C H Tham



party assignee.12 The reasoning underlying this conclusion is, however,rather terse. There being neither statutory nor public policy reasons torender such prohibitory clauses void,13 Lord Browne-Wilkinson read theexisting authorities14 as establishing that:

an attempted assignment of contractual rights in breach of a contractualprohibition is ineffective to transfer such contractual rights. I regard the law asbeing satisfactorily settled in that sense. If the law were otherwise, it woulddefeat the legitimate commercial reason for inserting the contractual prohibition,viz., to ensure that the original parties to the contract are not brought into directcontractual relations with third parties.15

But does equitable assignment have such an effect—that is, to bring theoriginal parties to the contract into direct contractual relations with thirdparties?16 More specifically, does it bring the original obligor into directcontractual relations with a third party assignee? In Warner Bros RecordsInc v Rollgreen Ltd, Sir John Pennycuick MR said not:

Where there is a contract between A and B, and A makes an equitable but not alegal [or more accurately, a statutory] assignment of the benefit of that contractto C, this equitable assignment does not put C into a contractual relation with B,and, consequently, C is not in a position to exercise directly against B any rightconferred by the contract on A. … [S]o long as the assignment remains equitableonly, C has no more than a right in equity to require A to protect the interestwhich A has assigned17

Perhaps Lord Browne-Wilkinson was implicitly overruling the positiontaken in Warner Bros. But that would require one to assume that anattempted equitable assignment would have the effect of a ‘transfer’ ofcontractual rights. So three questions present themselves. First, doesequitable assignment only operate to transfer contractual rights?18 Second,

12 Linden Gardens, above n 1, at 108.13 Ibid, at 106.14 Namely Helstan Securities Ltd v Hertfordshire County Council [1978] 3 All ER 262

(QB) in which Croom-Johnson J had held that a clause in a contract prohibiting theassignment of any benefits thereof was effective; Reed Publishing Holdings Ltd v King’sReach Investments (25 May 1983) CA transcript 121 which had dismissed an application tojoin an assignee as a party to the proceedings when the assignment was in breach of aprohibition against assignment; and Re Turcan (1888) 40 Ch D 5 (CA) which had proceededon the basis of a valid declaration of trust on the basis that a contractual restriction onassignment was valid: see Linden Gardens, above n 1, at 106.

15 Linden Gardens, above n 1, at 108.16 Lord Browne-Wilkinson mentions in passing that notice of the assignment was given to

the obligors in one of the two appeals: ibid, at 101. But it is plain from the emphasis placedon the dates of the assignment in both cases, leaving aside any question of notice, that theirLordships’ speeches were predicated on the assignments being equitable and not statutory: seen 3 above.

17 [1976] QB 430 (CA) 445.18 This is principally examined in Part II.A, below.

Equitable Assignment and Anti-Assignment Clauses 285



does it transfer contractual rights at all?19 Third, what does this tell usabout the position taken by their Lordships in Linden Gardens on theeffectiveness of anti-assignment clauses?20

A. Equitable Assignment Apart from Transfer

Given the commercial importance of assignment, it is a little surprising thatit has not received much academic scrutiny. Within the last hundred years,only a handful of treatises have examined its inner workings.21 Ofparticular note are Oshley Roy Marshall’s Assignment of Choses inAction22 and Joseph Starke’s Assignments of Choses in Action in Aus-tralia,23 and, much more recently, Marcus Smith’s Law of Assignment24

and Greg Tolhurst’s Assignment of Contractual Rights.25

For Marshall and Tolhurst, the answer to the first question would be,‘no, an equitable assignment does not only operate as a transfer ofcontractual rights’. Both take the view that equitable assignment of a chosein action may operate as a contract between assignor and assignee,26 aswell as a conveyance27 or transfer28 of the chose in action. Now, if it weretrue that equitable assignment operated by way of transfer of contractualrights, the obligor’s fear of being drawn involuntarily into direct contrac-tual relations with a stranger might well be real. But even if the contractbetween assignor and assignee causes some form of ‘transfer’ of the

19 This is principally examined in Part II.B, below.20 This is principally examined in Parts III.B and III.C, below.21 There have been, of course, numerous books which have dealt with the assignment of

choses in action as a single chapter within the much broader topic of contract law. But by andlarge these treatments have not delved very far beneath the surface of judicial authority,perhaps unavoidably, given the constraints of dealing with such a complex topic within theconfines of a more generalised discussion.

22 OR Marshall, The Assignment of Choses in Action (London, Sir Isaac Pitman & Sons,1950).

23 JG Starke, Assignments of Choses in Action in Australia (Sydney, Butterworths,Sydney, 1972).

24 M Smith, The Law of Assignment: The Creation and Transfer of Choses in Action(Oxford, Oxford University Press, 2007).

25 G Tolhurst, The Assignment of Contractual Rights (Oxford, Hart Publishing, 2006).26 So long as the equitable assignment is by means of a contract of assignment, of course.27 Marshall, above n 22, at 119, in relation to what Marshall terms ‘informal assign-

ments’ which are assignments operating ‘… neither by way of trust nor by way of contract…’.

28 Tolhurst, above n 25, at [3.25]–[3.26], [4.07]–[4.08]. As to what Tolhurst means by a‘transfer’ of a chose in action, see [3.10]. That conception incorporates notions of disposition,but is, ultimately, broad enough to encompass instances where there is no disposition of anyinterest in the chose at all: see text to and following n 54 below.

286 C H Tham



contractual rights in the chose in action assigned,29 why should theimpermissibility of the transfer affect the validity of the contract ofassignment?30

Even accepting the conveyance or transfer analysis as accurate, neitherMarshall nor Tolhurst suggests that equitable assignments may only takeeffect by means of a transfer. Marshall differentiates between three sensesof assignment in equity,31 namely an informal assignment,32 assignmentsby way of contract,33 and assignments by way of trust.34 Tolhurst, in turn,re-christens assignments by way of contract as the ‘remedial’ model ofassignment. Tracking Windeyer J’s analysis in Norman v Federal Commis-sioner of Taxation,35 he takes the view that this remedial model is ‘not atodds with the idea that an assignment involves a transfer’.36 So neitherMarshall nor Tolhurst denies that an assignment may operate as a contractbetween assignor and assignee (nor could they, given the many cases whichmake this very point).37 But as a contract between assignor and assignee,how might the obligor ever be brought into ‘direct contractual relationswith third parties’, as Lord Browne-Wilkinson feared?

First, to allow this fear of the obligor to override the contractual bargainstruck between assignor and assignee (where the assignment is for value)38

29 This assumption is questioned in Part II.B, below.30 For the contract of assignment is surely the cause of the purported transfer, and to hold

otherwise would be to allow the tail to wag the dog (assuming, of course, that the assignmentwas not intended to occur by way of gift).

31 Marshall, above n 22, at 80–99.32 Typified by the events in William Brandt’s Sons & Co v Dunlop Rubber Co Ltd [1905]

AC 454 (HL). Smith suggests that such ‘informal assignments’ are, in fact, instances where aconstructive trust has been imposed over the legal chose in action in question: Smith, above n24, at [6.12].

33 As noted by the Lord Chancellor in Wright v Wright (1750) 1 Ves Sen 409, 27 ER1111 (Ch) 412.

34 As noted by Cozens-Hardy LJ in Fitzroy v Cave [1905] 2 KB 364 (CA) 373. Marshalldifferentiates between the case where the assignor declares himself or herself to be a trustee(which would, in effect, be identical to the analysis undertaken above on the constitution ofan express trust over the chose in action), the case where the assignor of the chose transfers itto trustees on trust for the beneficiary-assignee, and the case where the debtor is directed bythe creditor to hold the chose in action in trust for the assignee.

35 (1963) 109 CLR 9 (HCA).36 Tolhurst, above n 25, at [3.24], further expanded at [4.05]–[4.06].37 Lord Browne-Wilkinson expressly accepted this to be true: Linden Gardens, above n 1,

at 108.38 The assignment under consideration in Linden Gardens was in the form of a deed, in

consideration for £1 paid to the assignors on execution of the deed: Linden Gardens (CA),above n 1, at 9 (as recounted by Staughton LJ); Linden Gardens (HL), above n 1, at 100 (asrecounted by Lord Browne-Wilkinson). In the other appeal, St Martins, the assignment of theassignor’s contractual rights against the building contractors was also made by way of a deed.That deed, however, also encompassed the assignment of the assignor’s proprietary interest inthe development in question. Both Staughton LJ and Lord Browne-Wilkinson made it clearthat consideration had been furnished in relation to the assignment of the proprietary interestin the development: St Martins (CA), above n 1, at 10; Linden Gardens (HL), above n 1, at101. But even though there is no express mention of consideration in support of the




is little different from the ill-fated attempt of the Dunlop Pneumatic TyreCompany to impose a minimum price floor on third party re-sellers of itsproducts, so forthrightly rejected by the House of Lords.39 And certainly, itis nowhere plausibly suggested that the effect of an anti-assignment clausemight be to prevent a disappointed contractual assignee from recoveringdamages from the assignor on account of losses arising from the failedcontract of assignment.40 To permit the anti-assignment clause to invali-date the contractual promise made by the assignor to the assignee wouldseem to be analogous, if not functionally equivalent, to the imposition of acontractual burden on a stranger to a contract. It is therefore difficult tosee how we can avoid applying the general principle at common law that acontract may not impose burdens on anyone who is not party to it. LordBrowne-Wilkinson himself observed in Linden Gardens that:

a prohibition on assignment normally only invalidates the assignment as againstthe other party to the contract so as to prevent a transfer of the chose in action:in the absence of the clearest words it cannot operate to invalidate the contractas between the assignor and the assignee and even then it may be ineffective onthe grounds of public policy.41

This extract gives rise to the implication that even with an effectiveanti-assignment clause, the contract of assignment still remains effective asbetween assignor and assignee. All the anti-assignment clause then does, itseems, is to prevent a transfer of contractual rights owed to the assignor bythe obligor to the assignee. And Lord Browne-Wilkinson makes this plainin the very next sentence, where he says:

If on the other hand Darling J purported to hold [in Tom Shaw v MossEmpires42] that the contractual prohibition was ineffective to prevent [theassignor’s] contractual rights against Moss Empire being transferred to TomShaw [the assignee], it is inconsistent with authority and was wrongly decided.43

Which is to say that Lord Browne-Wilkinson agreed that an anti-assignment clause only operates to invalidate the assignment as betweenthe assignor (who is in breach of the anti-assignment clause) and theobligor (whose obligation it is that has been assigned), and presumablythat it also prevents any possibility that the purported assignee might

assignment of the assignor’s contractual rights against the building contractor, this wasprobably unnecessary since this formed part of the same deed, and in all likelihood, theconsideration furnished by the assignees was intended to be in exchange for all the interests,proprietary or otherwise, which were to be furnished by the assignors under the deed.

39 Dunlop Pneumatic Tyre Co v Selfridge & Co Ltd [1915] AC 847 (HL).40 Indeed, Devefi Pty Ltd v Mateffy Pearl Nagy Pty Ltd (1993) 113 ALR 225 (FCA) 236

tells us otherwise.41 Above n 1, at 108.42 Above n 11.43 Linden Gardens, above n 1, at 108 (emphasis added).

288 C H Tham



obtain via the assignment any direct contractual rights against the obligorwhich were previously owed to the assignor. However, this begs thequestion whether equitable assignment operates to ‘transfer’ contractualrights, owed by an obligor to the assignor, to the assignee. I suggest that itdoes not.

B. Equitable Assignment as Transfer: An Impossibility?

(i) The Supporting Role of Equity

Leaving aside statutory assignments pursuant to the Law of Property Act1925,44 the received wisdom is that common law made no generalprovision for the assignment of choses in action.45 It just could not bedone. Any ‘transfer’ of the chose would, at common law, have to beeffected by novation (that is, a sort of surrender and regrant). It is alsoreceived wisdom that equity took a different view. In equity, it is said thatchoses in action are, in general, assignable.46 If, however, ‘transfers’ ofchoses in action were not possible at law (as noted above), and if, ‘equityfollows the law,’ how then may equitable assignment do what the lawcannot, namely ‘transfer’ choses in action?

Lord Macnaghten said in William Brandt’s Sons & Co v Dunlop RubberCo Ltd that:

[t]he [Supreme Court of Judicature Act 1873] does not forbid or destroyequitable assignments or impair their efficiency in the slightest degree.47

He went on to say that:

[w]here the rules of equity and the rules of the common law conflict, the rules ofequity are to prevail. … At law it was considered necessary that the debtorshould enter into some engagement with the assignee [such as a novation]. Thatwas never the rule in equity.48

There is little to quibble with in the first part of Lord Macnaghten’sstatement. But the latter part of his analysis requires more care. No specificauthority is cited, although he may have had in mind the history of

44 (UK) 15 & 16 Geo V c 20.45 Limited exceptions were made, eg, in relation to negotiable instruments and other

‘documentary’ choses in action.46 The editors of RP Meagher, Meagher, Gummow & Lehane’s Equity, Doctrines &

Remedies, 4th edn (Australia, Butterworths Lexis Nexis, 2002) 221 list the followingexceptions to equity’s benign attitude towards assignment: (i) bare rights of action; (ii)contracts involving personal skill and confidence; (iii) the salaries and pensions of certainpublic officers; and (iv) statutory rights which, expressly or impliedly, had been madeun-assignable.

47 Above n 32, at 461.48 Ibid.




jurisdictional disputes between the courts of common law and chancerythat was ultimately resolved in favour of the chancery. And even thoughsection 25 of the Supreme Court of Judicature Act 1873 provided forresolution of conflicts between rules of law and equity in relation to certainmatters, in particular the assignment of choses in action (in section25(6)),49 Lord MacNaghten’s observations ought surely to have beenqualified by section 25(11):

(11) Generally in all matters not herein-before particularly mentioned, in whichthere is any conflict or variance between the Rules of Equity and the Rules of theCommon Law with reference to the same matter, the Rules of Equity shallprevail.

Now, even though section 25(6) only provides for a statutory means forassignment of choses in action, it would seem that the assignment of chosesin action is, in consequence, a ‘matter … hereinbefore mentioned’ since itdoes address the question of assignment of choses in action. Such assign-ments are, therefore, matters which fall outside section 25(11): as particu-lar mention had been made in section 25(6) as to the assignment of chosesin action, there ought be no room for section 25(11) to operate so as topermit the rules of equity in relation to assignment of choses in action to‘prevail’ over those at common law.

There is, indeed, nothing in the doctrine of equitable assignment forvalue which requires any notion of ‘transfer’ of the chose in action. Facedwith a contractual agreement to assign a chose in action, equity, in itsauxiliary jurisdiction, would grant specific performance against theassignor to perform the terms of the contract with the assignee to assign bycompelling him or her to lend his or her name to any such proceeding thatthe assignee might wish to bring. There is no ‘transfer’.50 Rather, there isthe creation of a new equitable interest.

Tolhurst explicitly acknowledges much of the above:

[e]quity could not put in place rules for the transfer of legal rights. What it couldand did do was take the position that, if a person intends to assign a legal right,that person should in certain circumstances … be bound by that act. Thus,equity acting in personam attaches to the conscience of the assignor and forcesthe assignor to lend his or her name to the suit at law.51

49 (UK) 36 & 37 Vict c 66. Re-enacted as s 136(1) of the Law of Property Act 1925.50 As will be made clear, what the assignor does is not transfer any pre-existing rights.

Rather, he or she creates a new right, assuming a new obligation recognised in equity owed tothe assignee. See also Part II.B below.

51 Above n 25, at [4.05].

290 C H Tham



Had the assignees in both appeals in Linden Gardens sought specificperformance of the contract of assignment,52 compelling the assignor tolend its name for the purposes of their bringing an action against theobligors to the choses in action assigned, no transfer of property wouldhave been involved. At best, equity would recognise the ‘assignee’ ashaving some interest in the chose in action assigned, and that new interestis patently not the same as the chose in action assigned. Most obviously thenew interest is not primarily directed at the obligor to the chose in action.The assignee’s interest arises out of the ‘assignment’ and is directed at the‘assignor’, against whom the order of specific performance is made. Andthe subsequent procedural changes which short-circuit the need to bringseparate equitable proceedings against the assignor do not change thesubstantive logic underlying such assignments. So on this ‘non-transfer’ or‘contractual’ view of equitable assignment, assignee and obligor arepatently not brought into direct contractual relations with each other, andso the policy deployed by Lord Browne-Wilkinson to underpin his view inLinden Gardens ought not to have been of any concern.

For there to be an effective ‘transfer’ at common law, a novation isnecessary.53 And though the provisions for a statutory assignment leftuntouched the position in equity for assignments of a chose in action, suchequitable assignment cannot ‘prevail’ over the common law requirementfor a novation. So an equitable assignment cannot place the assignee in thesame position as he or she would have been in had there been a novation.It may allow the equitable assignee to achieve much of what might haveoccurred had there been a novation, but complete congruency is notpermissible, since to allow an equitable assignee to be in the position as ifthere had been a novation at law without having had to satisfy thecommon law’s requirements would be to contradict the Judicature Act.

(ii) A Minimalist Usage of the Language of ‘Transfer’

As set out above, an equitable assignment may operate as a contractualpromise between assignor and assignee, in particular the promise by theassignor to lend his or her name to the assignee for the purposes ofbringing legal proceedings against the obligor. The ubiquity of the verb ‘totransfer’ is, however, difficult to escape. So perhaps it is appropriate tocontinue using that verb, but only if we are careful about what is being‘transferred’.

52 Presumably, this was not sought because the assignors in both appeals were only toohappy to assist. But the point is that specific performance was available had it been sought.

53 Or, following the enactment of the Judicature Act, compliance with the requirements toeffect a statutory assignment pursuant to s 25(6) (or today s 136(1) of the Law of PropertyAct 1925).




I suggest in this article that an assignor ‘transfers’ a chose in action byhanding over control of his or her freedom to decide whether or not torelease the obligor from the contractual obligations under the chose inaction assigned. This right to grant a release, of course, encompasses itscorollary: the right to bring legal proceedings on the chose in action shouldthere be a dispute as to the degree to which the obligor had completelyperformed the obligations under the chose.

The limited nature of equitable assignment is, in fact, explicitly recog-nised by Tolhurst:

one might logically argue that although not all dispositions of rights involvetransfers, a transfer requires a disposition. Such a restricted notion of title‘transfer’ would not capture an equitable assignment of a legal right which at ananalytical level merely creates and vests in the assignee an equitable interest. Theassignor does not dispose of an equitable interest because when, one personholds ‘the whole right of property,’ no distinction is drawn between legal andequitable interests. Equitable interests are ‘engrafted’ or ‘impressed’ upon legalinterests rather than ‘carved out of’ them.54

To fit this fact pattern, whereby an equitable interest in the chose in actionis created by virtue of what takes place between the assignor and theassignee, Tolhurst stretches the meaning of ‘transfer’ to include this case.55

But, with respect, if we wish to use the word ‘transfer’ in the context of anequitable assignment of a chose in action, we need to take care to recognisethat it is a transfer that does not involve any disposition.

(iii) Is There Really Nothing More? Competing ‘Maximalist’ Views

Commentary on the question as to what is transferred by an equitableassignment of a legal chose appears to assume that the subject matter ofthe transfer is something more than just the right to bring legal proceed-ings. Smith observes that:

[f]irst, and most narrowly, what is assigned could be no more than a right to suein the name of the assignor. If this is what an equitable assignment of a legalchose is, it is remarkably similar in concept to the common law use of a power ofattorney to enable an assignee of a chose to enforce the rights of the assignor.56

While acknowledging that there are cases57 which hold that an equitableassignment merely gives the assignee the right to sue in the assignor’s name,

54 Above n 25, at [3.11] (citations omitted).55 Ibid.56 Smith, above n 24, at [6.07].57 Smith cites Winch v Keeley (1787) 1 TR 619, 99 ER 1284 (KB) 623; Crouch v Credit

Foncier of England (1873) LR 8 QB 374, 380; De Pothonier v De Mattos (1858) El Bl & El461, 120 ER 581 (KB) 467; The Wasp (1867) LR 1 A & E 367 (HC Adm) 368; Walter &Sullivan Ltd v J Murphy & Sons Ltd [1955] 2 QB 584 (CA) 588–9; Three Rivers District

292 C H Tham



Smith notes that they ‘do not go so far as to state that this is the only rightthat the assignee acquires by virtue of the assignment’.58 Smith posits that asecond, alternative view is preferable, that is:

the assignment could effect the transfer of the beneficial interest in the chose,leaving the assignor with nothing but the bare legal title. This implies asubstantive effect on the assignor’s rights in that the beneficial interest in thechose is separated from the legal title. In turn, this implies the creation of atrust.59

Smith argues that this second view is preferable because the first narrowconception of assignments would not be able to explain how, afternotification of the assignment, the obligor may not be discharged from hiscontractual obligation unless he performs to the assignee.60

Tolhurst is of roughly the same opinion:

Historically, for an equitable assignee of a legal right to enforce the legal rightthe action had to be brought by the assignee in the name of the assignor. If theassignor refused to allow its name to be used the assignee could, if theassignment was for valuable consideration, file a bill in equity and, upon givingan indemnity in costs, obtain an injunction allowing the assignor’s name to beused in a suit at law to recover the assigned debt.…

On the above analysis the equitable assignment of a legal right merely providedthe assignee with a remedy against the assignor, that is, the assignment operatedonly between the assignor and assignee. What is obtained or ‘assigned’ from theassignor appears to be no more than a right to sue in the name of the assignor. …The effect of this contract analysis is that equity forces the assignor to performits promise rather than leave the assignee with a remedy in damages. It isimportant to note what that promise is. Initially, the assignor had agreedimmediately to assign a legal right to the assignee. Such an agreement wasincapable of being performed at law61 and therefore it cannot be the case thatequity would uphold such contracts for assignment only if they were capable ofbeing the subject of an order for specific performance. It appears that equityimplied a promise that the assignor would lend its name to any suit against theobligor and held the assignor to this promise.62

Council v Bank of England [1996] QB 292 (CA) 299: see ibid, at 145 fn 16. Making muchthe same point, Tolhurst cites Hammond v Messenger (1838) 9 Sim 327, 59 ER 383 (Ch):above n 25, at 69 fn 19. He also cites Long Leys Co Pty Ltd v Silkdale Pty Ltd (1991) 5Butterworths Property Reports 11,512, 11,518; Showa Shoji Australia Pty Ltd v Oceanic LifeLtd (1994) 34 NSWLR 548 (SC) 561; Corin v Patton (1990) 169 CLR 540 (HCA) 576;Neave v Neave [1926] Gaz LR 254, 256: above n 25, at 70 fn 20.

58 Above n 24, at [6.07] (emphasis in original).59 Ibid (emphasis in original).60 Ibid, at [6.11]. This view might, however, require re-examination in light of the

discussion of the cases relied upon in support: see Part II. C (iii), below.61 In the absence of any statutory assignment pursuant to the Judicature Act or the Law

of Property Act 1925. Equitable assignment, of course, predates both statutes.62 Above n 25, at [4.05].




However, in his opinion, there were two problems with this analysis. First,there would be no need to resort to the language of assignment and vestedrights.63 Second, it would prevent the recognition of voluntary assign-ments.64 Both problems would be obviated if, however, one accepted thatequitable assignments were truly transfers, whereby the assignee would be‘vested with the ownership of rights existing between the assignor and theobligor by virtue of the contract between the assignor and obligor’.65 Andthis state of affairs would arise, in Tolhurst’s opinion, because:

equity restated or progressed the remedial model so that the transfer was giveneffect to by reason of the remedies equity provided.

Here, equity does not simply bind the conscience of the assignor and act inpersonam; rather, because the conscience of the assignor is bound equity fastensupon the subject property itself.…

[A]n equitable interest is created and vested in the assignee, and that interest isthe beneficial ownership of the legal right which is the subject of the assignmentand the assignee is treated (in equity) as being owed the obligation. That is, inless precise terms (and in respect of contractual rights), the assignee obtains aninterest in the contract that exists between the assignor and obligor. This isreferred to here as the assignment analysis.66

Broadly speaking, therefore, both Smith and Tolhurst deny that anequitable assignment only takes effect by reference to the right to compelthe assignor to lend its name for the assignee to bring legal proceedingsagainst the obligor. Both suggest that in equity, some form of propertyinterest is created and vested in the assignee by virtue of the equitableassignment, pointing out various problematic issues if such were not thecase. A few observations may be made in reply.

First, there is no inherent inconsistency between the view that anequitable assignment involves only a ‘transfer’ of the right to grant arelease (and, its corollary, the right to decide to bring legal proceedings onthe chose in action) and the view that some equitable interest is created inthe chose in action. That being the case, it is perfectly conceivable that thesubject matter of the constructive trust67 or the equitable interest68 in thechose, as advocated by Smith and Tolhurst, merely relates to the right to

63 Ibid, at [4.06].64 Ibid, at 70 fn 25. Presumably, this is because specific performance has no role in a case

of a completed voluntary gift of a chose in action (there being no contract to specificallyenforce). Nor would it have any role in relation to incomplete gifts (since equity does not lendits assistance to volunteers).

65 Ibid, at [4.06].66 Ibid, at [4.07].67 On Smith’s view.68 On Tolhurst’s view.

294 C H Tham



grant a release and does not modify the terms specifying what is toconstitute complete contractual performance by the obligor.

Second, though the focus of this article has been on equitable assign-ments for value, and on the specific enforceability of the contractualpromise to assign, acceptance of this view does not preclude the possibilitythat some different basis may be relied upon to explain the phenomenon ofvoluntary equitable assignment, perhaps along the lines suggested byTolhurst.69

Third, one ought not lose sight of contract principles pertaining todischarge by performance. Tolhurst distils the rules regulating inter vivosassignment of contractual rights as follows:

the assignment of contractual rights is made up of a number of rules whichoverlap, make little sense as statements in their own right and appear to lack anygeneral underlying and unifying principle. These rules are as follows:

1. An assignor can assign no greater right than it has nor can an assignee obtaina right greater than that held by the assignor.

2. Only non-personal contractual rights may be assigned.

3. It is not possible by assignment to increase or vary the obligations or burdensof an obligor.

4. It is possible to assign only rights and not obligations.

5. After receiving notice of an assignment, the obligor may not do anything todiminish the rights of the assignee.

6. An assignee can be in no better position than the assignor was prior to theassignment.

7. An obligor should be no worse off by virtue of an assignment.

8. An assignee takes subject to the equities.70

Tolhurst seeks to explain and rationalise these rules in Anglo-Australianlaw by proving two things. First, that all assignments, whether legal(statutory) or equitable do not just involve a transfer of rights; rather, forTolhurst, assignment is transfer.71 Second, that his Rules 5 to 8, ‘to theextent that they focus on the obligor/assignee relationship rather than theassignor/assignee relationship, are explicable on this basis rather than theprinciple of transfer’.72 In general, however, because Tolhurst begins with aconception of assignment as a form of transfer, the rule of nemo dat looms

69 Alternatively, see CH Tham ‘Careless Share Giving’ (2006) 70 Conveyancer andProperty Lawyer 411, 421ff.

70 Above n 25, at [1.01].71 Ibid, at [1.02].72 Ibid.




large throughout his analysis. Thus, he takes the view that his Rules 1, 3, 6and 8 are derived from the fundamental idea that a transferor can onlytransfer what he or she has at the time of transfer, and not anything more.This analysis is useful in certain contexts. However, for the purposes of thisarticle, Tolhurst’s starting points in relation to his treatment of Rule 3, aslisted above, require closer scrutiny.

C. Equitable Assignment, (In)Variability of Contract Terms andDischarge by Performance

(i) Difficulties with the Maximalist Views

In relation to his Rule 3, Tolhurst makes the following assertion:

[T]his rule is a clear adoption of the nemo dat rule. If the assignor cannot assigna right different from or greater than the one vested in him or her, then the effectof that assignment must be that the correlative obligation of the obligor is alsonot capable of variation by reason of the assignment.73

He then explains that:

[it] may be noted that ‘variation’ here concerns variations of legal obligationsrather than factual changes to performance. So long as the obligor is being askedto perform for the assignee the exact same legal obligation as that promised tothe assignor, then the obligor cannot complain that there has been a variation toits obligation. It is accepted that there may be some increased inconvenience infact by reason of an assignment.74

Both of these statements may be disputed. It is not obvious that the rule asto the non-variability of an obligor’s contractual obligations derives fromthe nemo dat rule, for that non-variability plainly exists even where there isno question of assignment. Unless a contract expressly or impliedly permitsthe parties to a contract to modify the obligations undertaken by one partyto the other, those obligations must be performed strictly in conformitywith the agreed terms in order to discharge them by performance.Otherwise, the obligations as set out in such terms would be breached.This is really the result of the contractual principle that, unless otherwiseprovided for, ‘[t]he general rule is that a party to a contract must performexactly what he undertook to do’.75 It is not that assignment is incapableof effecting a variation because of nemo dat; rather, unless otherwiseprovided for, contractual provisions are inherently invariable.

73 Ibid, at [1.03].74 Ibid, at 6.75 HG Beale (ed), Chitty on Contracts, 29th edn (London, Sweet & Maxwell, 2004) vol 1

[21–001].

296 C H Tham



Of greater concern is Tolhurst’s second statement, requiring us todistinguish between ‘variations of legal obligations’ and ‘factual changes toperformance’. The former, he agrees, is impermissible and may not occureven where there has been a valid assignment. He reasons, however, thatthe latter is an inevitable and judicially sanctioned outcome. But how dowe distinguish one from the other? To permit a ‘factual change toperformance’ is to permit a ‘variation of legal obligation.’ They are oneand the same because the question whether a factual performance hassatisfied the requirements of the contract is determined by comparing whathas in fact been done with what in law is required, the latter beingdetermined through applying an appropriate construction of the relevantcontractual terms. This is borne out by Tolhurst v The Associated PortlandCement Manufacturers (1900) Ltd.76

In that case, Alfred Tolhurst (the obligor) owned land at Northfleet inKent containing extensive chalk quarries. He sold a small piece of this landto the Imperial Portland Cement Company Ltd (the obligee) whichconsolidated it with another piece of land so as to establish a factory toproduce Portland cement. In January 1898, the obligor entered into acontract with the obligee to the effect that he would:

for a term of fifty years, to be computed from the 25th day of December, 1897,or for such shorter period (not being less than thirty-five years) as he shall bepossessed of chalk available and suitable for the manufacture of Portlandcement, and capable of being quarried and got in the usual manner above waterlevel, supply to the company, and the company will take and buy of the said[obligor] at least 750 tons per week, and so much more, if any, as the companyshall require for the whole of their manufacture of Portland cement upon theirsaid land.77

They contractually agreed that 1s 3d per ton would be paid in cashmonthly for the chalk, and the average monthly payment for any year after1898 was not to be less than £188.

As permitted by its constitutive documents, the obligee was voluntarilyliquidated in 1900. Prior to that, it sold its entire undertaking, includingthe contract with the obligor, to the Associated Portland Cement Manufac-turers (1900) Ltd (the assignee) via a deed of assignment.78 The obligortook no part in the liquidation of the obligee, but appears to have suppliedsome chalk to the assignee, asking 2s per ton for it, on the basis that the1898 contract had been brought to an end when he was given notice of theliquidation and sale of business. When the assignee declined to pay at thenew rate, insisting instead that it was entitled to the benefit of the 1898

76 [1903] AC 414 (HL) [Portland Cement].77 Ibid, at 418.78 This is made clear in the judgment of Collins MR in the court below: Tolhurst v

Associated Portland Cement Manufacturers (1900) Ltd [1902] 2 KB 660 (CA) 665.




contract and to pay only 1s 3d per ton, the obligor brought an actionagainst the assignee for the difference and for a declaration that he was nolonger under any obligation to the obligee. The assignee and obligeecross-claimed for a declaration that the 1898 contract was still valid andbinding, and that the assignee and the obligee were entitled to supply onthe original terms.79

Mathew J at first instance found for the obligor on both the claim andthe cross-claim. To his mind, it was:

perfectly clear that [the assignee was] endeavouring to impose upon [the obligor]a contract into which he never entered, and that he [was] entitled to say that hedid not make that contract. … Whenever, by reason of change of business orchange of parties, the result of an assignment would be to impose upon one ofthe contracting parties a greater liability than he ever intended to assume, thecontract [could not] be assigned.80

The Court of Appeal reversed Mathew J’s decision, for the followingreasons. First, it made no difference to the obligor whether payment of theprice for chalk delivered pursuant to the 1898 contract was made person-ally by the obligee or some other party: that obligation was non-personal(in other words, payment of the price could be vicariously performed).81

Second, the 1898 contract had not been repudiated, notwithstanding theobligee’s liquidation and the sale of its business to the assignee.82 Third, the1898 contract only provided for a minimum level of supply—it did notprovide for an upper limit of supply by reference to the expected scale ofproduction. So there was no merit to the obligor’s claim that he would, ineffect, be put under a greater burden in light of the assignee’s scale ofbusiness compared with that of the obligee.83 Therefore, the obligor wasnot entitled to succeed on his claim for payment for chalk sold anddelivered at a rate of 2s per ton, nor was he entitled to succeed on hisapplication for a declaration that the 1898 contract was no longer bindingupon him.

By a majority, the House of Lords upheld the decision of the Court ofAppeal except on one critical point.84 Delivering the leading judgment,Lord Macnaghten85 took the view that although the 1898 contract onlymade express reference to the obligor and obligee, and made no express

79 The particulars of the litigation and relief sought by the parties are culled from theheadnote in the report of the first instance judgment of Mathew J: Tolhurst v AssociatedPortland Cement Manufacturers (1900) Ltd [1901] 2 KB 811, 813.

80 Ibid, at 816–17.81 Tolhurst v Associated Portland Cement Manufacturers (1900) Ltd, above n 78, at 672,

679.82 Ibid, at 672–4, 678–9.83 Ibid, at 673, 680.84 Lord Robertson was the sole dissenting judge: Portland Cement, above n 76, at 421–2.85 With whom Lord Shand and the Earl of Halsbury LC agreed (albeit reluctantly).

298 C H Tham



provision for the assigns or successors in title of either, given that it was tolast for at least 35 if not 50 years, a plain and literal reading of the contractwould not do. Rather:

when it is borne in mind that the [obligee] must have been induced to establishits works at Northfleet by the prospect of the advantages flowing fromimmediate connection with [the obligor’s] quarries, and that the contract insubstance amounts to a contract for the sale of all the chalk in those quarries byperiodical deliveries … , it is plain that it could not have been within thecontemplation of the parties that the company would lose the benefit of thecontract if anything happened to [the obligor], or that [the obligor] would losethe benefit of the market which the contract provided for him at his very door inthe event of [the obligee] parting with its undertaking, as it was authorized to doby its memorandum.86

Commercial reality required a more nuanced construction of the contract.Lord Macnaghten construed the 1898 contract as if it contained aninterpretation clause saying that references to the obligor were to includehis heirs, executors, administrators and assigns, owners and occupiers ofthe Northfleet quarries, and that references to the obligee were to includeits successors and assigns, owners and occupiers of the cement works atNorthfleet.87 In consequence, the obligor was not entitled to recoverpayment for the chalk delivered to the assignee at the 2s rate. Nor was theobligee a necessary or proper party to the litigation.88

The key question was whether the obligor, Tolhurst, was bound tocontinue performing the obligation to deliver chalk, even though theobligee, the Imperial Portland Cement Co, was, practically speaking, nomore.89 In answer, Lord Macnaghten adopted a construction of thecontract that expanded the meaning of the named obligor and obligee toinclude their successors and assigns.90 It naturally followed that theobligor’s principal objection to having to continue to perform his part ofthe contract, by reason of its impossibility of performance, had to fail.

86 Portland Cement, above n 76, at 419. See also Nokes v Doncaster AmalgamatedCollieries [1940] AC 1014, 1020, 1039.

87 Ibid, at 420.88 Ibid, at 421. Lord Macnaghten did not explain why this was the case, but an

explanation is hazarded in the text below at n 92.89 Lord Macnaghten remarked that though the joinder of the assignor would typically be

expected, in his view, this was unnecessary where the assignor was, ‘a mere name … withoutany executive or board of directors …’: ibid, at 420–21.

90 Tolhurst accepts that the obligor in Portland Cement ‘had contracted to supply theneeds of any person taking over the cement works who took an assignment of the right’:above n 25, at [6.121]. But, taking the position that personal obligations are non-assignable,he seems to limit the significance of the construction point in Portland Cement to the questionof whether the obligor was under a personal obligation to supply chalk to the assignors, theImperial Portland Cement Co: above n 25, at [6.70].




(ii) Portland Cement as a Case of Construction

Portland Cement was not decided on assignment principles per se, but on acommercially sensible reading of the 1898 contract. Had the assignee beenin existence on 12 December 1898, when the contract was formed, wewould have been able to stop there, for on Lord Macnaghten’s construc-tion of the contract the contractual obligation of the obligor would then beowed not only to the obligee, but also to such person as might fall withinthat extended construction, such as the assignee. We would then besquarely within the realm of joint contractual promisees. But the facts ofPortland Cement prevent us from applying that simple explanation. Even ifthe 1898 contract was read to include references to the obligee and itsassigns or nominees, as at the formation of that contract on 12 December1898, this particular assignee had not come into existence: it was onlyincorporated some time in 1900.91 Therefore, it could not be said that, asat 12 December 1898, any contractual bargain had been constitutedbetween the assignee and the obligor. Some other legal mechanism wasrequired to fill the gap, but we do not need assignment to do the job.

The clue lies in Lord Macnaghten’s explicit statement that the obligeewas not a necessary party to the action.92 A simple explanation could befound in the application of the doctrine of offer and acceptance, inparticular, open offers. Approaching the 1898 contract from this perspec-tive, and given the construction placed upon it by Lord Macnaghten, it isnot difficult to take the view that the clause setting out the obligor’scontractual responsibilities could be construed as an open offer to theobligee’s assigns. The significance of the deed of assignment, naming theassignee, was that it enabled the assignee to fall within that category ofpersons to whom the obligor’s offer to be bound was open. That offer wasaccepted when the assignee failed to reject that first delivery of chalk by theobligor after having been given notice of the assignment.

On this analysis, the sale of the business undertaking to the assignees didnot ‘transfer’ the rights to performance of the 1898 contract: it onlyoperated to allow the assignee to qualify as an offeree. Thus, PortlandCement is not a case about assignment as ‘transfer’ of contractual rights.93

This should come as no surprise, for Lord Macnaghten’s opening words in

91 This is clearly stated to have been the case in the headnotes of the reports of PortlandCement at first instance, above n 79, at 812, and in the Court of Appeal, above n 78, at 662.

92 Portland Cement, above n 76, at 421.93 This also explains, rather simply, why Lord Macnaghten took the view that the

assignees would have been contractually obliged to order and purchase chalk from Tolhurstto the extent and at the rate stipulated in the 1898 contract, had the shoe been on the otherfoot: see Portland Cement, above n 76, at 420. Assignment theory cannot explain suchassignment of burdens without resorting to rather complex theorising as to how, in somecases, burdens pass alongside benefits: see, eg, Tolhurst, above n 25, at [6.101]–[6.135], inparticular, [6.116]–[6.121].

300 C H Tham



his speech in Portland Cement warned us as much: that though the casemight be of great importance to the parties, from a legal point of view thiscase was of no importance at all.94

Lord Macnaghten’s reading-in of the additional interpretation clause iscentral to his analysis, for had it not been read into the contract, it isarguable that the ‘assignment’ would have failed even if all statutoryrequirements had been satisfied.95 As a statutory assignee, the assigneewould have had transferred to it any remedy which the obligee might havebeen able to assert against the obligor once notice in writing of theassignment had been given. But without the interpretation applied by LordMacnaghten, on a plain reading of the express terms of the contract, wouldthe obligor have been in breach of the contract if he refused to deliver anymore chalk? Of course not, since the named party to whom deliveries wereto be made had been liquidated and, presumably therefore, it would nolonger have been possible for the obligor to perform to the letter of thecontract. This is the corollary of the principle of discharge by fullperformance.96 Taking this principle seriously, if contractual performancerequires performance to be rendered to a particular named person, deliveryof goods or payment of that sum of money to another does not amount toprecise performance and does not, therefore, discharge the contract.97

Discharge, if it occurs, occurs because of some other doctrine.

(iii) Discharging a Debt by Making Payment to a Third Party

Admittedly, there is some authority for the proposition that in relation tothe payment of debts, though the amount, time and place of payment areinvariable, the party to whom the payment is to be made may be varied atthe option of the creditor-obligee. Tolhurst provides the following example:

if X owes A a debt of £100 payable at a certain place on a certain date, althoughA may be able to assign its right to the debt to a third party, that assignmentalone cannot force X to pay the debt at another place or on another date.98

94 Above n 76, at 416–17.95 At that point in time, the relevant provision would have been found in section 25 of the

Judicature Act.96 If authority is needed, one might start by considering Sir Thomas Plumer MR’s

observation in Goldsmid v Goldsmid (1818) 1 Swans 211, 36 ER 361 (Ch) 219 (in relation toa specialty debt): ‘Satisfaction supposes intention; it is something different from the subject ofthe contract, and substituted for it; … but with reference to performance, the question is, Hasthat identical act which the party contracted to do been done?’ Allan Farnsworth put it thisway: ‘If a duty is fully performed, it is discharged. … The converse is equally clear. Nothingless than full performance operates as a discharge.’: EA Farnsworth, Farnsworth on Contracts(Boston, Little, Brown and Co, 1990) vol 2 § 8.8.

97 Unless we draw the distinction proposed by Tolhurst between ‘legal’ and ‘factual’variations: see the quotation at text accompanying n 74 above.

98 Above n 25, at [6.93].




For this, he principally relies99 on the academic authority of Corbin onContracts.100 Both Tolhurst and Arthur Corbin leave open the possibilitythat the assignment may permit ‘variations’ in the identity of the party towhom X is to make payment in order to be discharged from his or herobligation of debt. Further, there is some old authority on which theeditors of Chitty on Contracts rely in support of the proposition that if acreditor-obligee requests the debtor-obligor to pay the debt to a third party‘such a payment is equivalent to payment direct to the creditor, and is agood discharge of the debt’.101 But those authorities, namely the old casesof Roper v Bumford102 and Page v Meek,103 require careful handling.Roper v Bumford is really a case of equitable set-off by agreement, andPage v Meek sets up no independent rule as to payment, but merely acceptsas a point of pleading that payment to a third party at the direction of acreditor may be made as a special plea of payment. But from the face of thereport, it would appear that the legal rationale for such pleading rested onthe doctrine of accord and satisfaction. So it is significant that Chitty onContracts merely states that the payment to a third party at the direction ofa creditor is equivalent to payment direct to the creditor so as to dischargethe debt. Though it is true that there may be a good discharge in suchcases, the reason for such discharge does not lie in any equitable assign-ment per se, but by reference to other doctrines: set-off in one case, andaccord and satisfaction in the other.

Similar treatment may be applied to some other cases that are often citedin support of the proposition that an obligor who makes payment orcompletes performance to the obligee-assignor instead of the assignee,following notice of an assignment, does so at his or her peril.104 Yet thecentral case, Brice v Bannister,105 may be read as being premised either onthe lack of mutuality between the obligor and the assignee necessary to

99 Ibid, at fn 754. Tolhurst also cites RA Brierley Investments Ltd v Landmark Corp Ltd(1966) 120 CLR 224, 231–2, 236; O Lando et al, Principles of European Contract Law(Dordrecht, Kluwer, 2003) Arts 11:306(1)–(2); Unidroit Principles of International Commer-cial Contracts (2004) Arts 9.1.8, 9.1.3. The relevance of these authorities, particularly thelatter two, is somewhat limited. The relevance of RA Brierley Investments Ltd v LandmarkCorp Ltd is also somewhat obscure as it deals with the rather distinct issue as to whether anentity that acquired shares in a company after a takeover offer had been made was entitled toaccept such an offer.

100 AL Corbin, Corbin on Contracts (St Paul, West Publishing, 1951) vol 4 § 868.101 Beale, above n 75, at [21–042].102 (1810) 3 Taunt 76, 128 ER 31 (CP).103 (1862) 3 B & S 258, 122 ER 98 (KB).104 See E Peel, Treitel on the Law of Contract, 12th edn (London, Sweet & Maxwell,

2007) 725–6, fn 95. This proposition was also doubted by Simon Brown LJ in DepositProtection Board v Barclays Bank Plc [1994] 2 AC 373 (CA) 382, although the decision ofthe Court of Appeal was ultimately reversed by the House of Lords on other grounds: [1994]2 AC 367 (HL).

105 (1878) LR 3 QBD 569 (CA).

302 C H Tham



enable the obligor’s advance payments to be statutorily set-off106 againstthe debt arising from work done on a contract to build a boat, or on thebasis of an insufficient degree of connection between such debts as toenable one to be set-off against the other in equity. As for Jones vFarrell,107 it is not too difficult to read it as a case where there was animplicit finding that the contractual debt had been discharged not byperformance to a third party at the behest of the creditor, but by accordand satisfaction, since the debtor in that case promised, in response to anorder by the creditor to do so, that he would pay the third party debtorsuch sums as would become due under the agreement which was thesubject of the assignment by the creditor. As for the remaining cases of exparte Nichols108 and Durham Bros v Robertson,109 the former was onlyconcerned with the issue of whether an equitable assignment of futuredebts executed prior to the bankruptcy of the assignee but after theoccurrence of an act of bankruptcy was good against the assignee’s trusteein bankruptcy. The latter was concerned with whether the assignmentbefore the court was a statutory or an equitable one. Like many of thecases encountered in the course of this analysis, none of these cases standas solid authority for the point for which it is commonly cited.

(iv) Equitable Assignment, the Invariability of Contract Terms andDischarge by Performance: a Synthesis

As this article has tried to demonstrate, the attempt to cast equitableassignment as a form of transfer creates unworkable inconsistencies if theconcept of ‘transfer’ is applied indiscriminately. Cases such as Roper vBumford,110 Page v Meek,111 Brice v Bannister,112 or Jones v Farrell113 donot compel one to accept that payment to the original creditor-obligeemust be treated as being of no effect so far as discharge of the obligor’spayment obligation is concerned. The final outcomes in those cases areexplicable not by reference to some overriding principle peculiar toequitable assignment, but rather by reference to a variety of other legal

106 Pursuant to s 13 of the Insolvent Debtors Relief Act 1729 (UK) 2 Geo II c 22 asamended by the Debtor’s Relief Act 1735 (UK) 8 Geo II c 24 [the Statutes of Set-off]. InEngland, although the Statutes of Set-off were repealed in 1879 by s 2 of the Civil ProcedureActs Repeal Act, as the Court of Appeal has made clear, s 4(1)(b) of that statute preserved therights of statutory set-off conferred by the Statutes of Set-off: see Glencore Grain Ltd v AgrosTrading Co [1999] 2 Lloyd’s Rep 410 (CA) 417.

107 (1857) 1 De G & J 208, 44 ER 703 (Ch).108 (1883) 22 Ch D 782 (CA).109 [1898] 1 QB 765 (CA).110 Above n 102.111 Above n 103.112 Above n 105.113 Above n 107.




doctrines. The availability of means other than precise performance asdefined in the contract between obligor and assignor to effect a dischargedoes not logically compel the conclusion that discharge by means of preciseperformance is no longer possible, and to insist on this may be excessivelyreductionist. This is entirely consistent with the analysis of PortlandCement set out above. For if equitable assignment varied the originalobligation of payment so as to change the identity of the party to whomthe contractual performance was owed, Lord Macnaghten’s efforts toextend the meaning to be given to the term identifying the obligee in thecontract would have been superfluous.

As noted above, it is axiomatic that a contractual obligation may only bedischarged by precise and exact performance. The degree of exactitude willdepend on the construction of the contractual obligation in question.However, where the party to whom delivery or payment is to be made isspecifically named, unless the contract, as a matter of construction, permitssuch delivery or payment to be made to an entity other than that namedparty, it is arguable that such performance would be an unwarrantedvariation of the contract—the performance would be neither precise norexact—and that obligation would be breached. This problem cannot beresolved by reference to assignment principles alone, and hence the needfor Lord Macnaghten to have interpreted the contract in Portland Cementto include references to the obligee’s assigns. In other contexts, the rules asto set-off, accord and satisfaction, or, perhaps, agency,114 may have a roleto play in appropriate circumstances so as to effect a discharge of thecontract otherwise than by precise performance. The point is that equitableassignment does not address the question of whether a contractualobligation may be discharged by performance to the assignee. Suchdischarge, if it occurs, arises as a result of other legal doctrine. Equitableassignment, therefore, does rather less than is commonly claimed.

III. IMPLICATIONS OF THE MINIMALIST VIEW

A. Non-Assignability of Personal Obligations

To some extent, the points made above are not new. Writing in 1926 onthe topic of assignment of contract rights, Corbin argued in favour ofabandoning the language of ‘alienability of choses in action’:

In continuing the discussion [on the alienability of choses in action] our first stepshould be to abandon altogether the term ‘chose in action.’ Its linguisticconstruction is faulty, in that its individual words lead one to think of something

114 Alluded to by Chitty LJ in Durham Bros v Robertson, above n 109, at 770.

304 C H Tham



very different from that which the expression as a whole now denotes. There isno ‘chose’ or thing or res. There is a right (or claim) against some person.115

This is, however, still a little too narrow. For if anything is ‘transferred’ orhas been made the subject of an equitable interest that is vested in the‘assignee’, it is the right to release the obligor from his or her contractualobligation.116

Realising what is ‘transferred’ in an equitable assignment of a contrac-tual chose in action has certain implications. One of them requires us tore-examine Tolhurst’s Rule 2, that ‘only non-personal obligations can beassigned’. Given that an equitable assignment may operate by way ofconstitution of a trust, this rule is overstated. As Lightman J pointed out inDon King Productions Inc v Warren, there ought to be:

no objection to a party to contracts involving skill and confidence [such as apersonal obligation] or containing non-assignment provisions from becomingtrustee of the benefit of being the contracting party as well as the benefit of therights conferred.117

Agreeing with Corbin, the position in this article is that the same may besaid where the equitable assignment operates otherwise than by constitu-tion of a trust. Where it is only the right to grant a release that is‘transferred’ by the equitable assignment, why should the fact that the legalproceedings relate to a ‘personal’ obligation invalidate such transfer?Corbin offers the following illustration:

A contracts with B to act as B’s valet. Surely, it will be said, B’s right is sopersonal that it cannot be assigned. But no, the contrary is believed to be correctalthough no decision pro or con has been seen by the writer. By this statement itis not meant to say that the character of the service can in any way be changedby assignment. The right of B is that A shall act as B’s valet, not that A shall actas valet for whom it may concern. Anyone ought to know that serving as valet toa cross, ill, miserly, old curmudgeon is not the same performance as serving ahealthy, happy-go-lucky, generous, young prince. Therefore, when B assigns hisright against A, he must assign it as it is. He cannot by assignment to C create inC a right that A shall act as C’s valet. That would be a different right to adifferent performance. But B can assign to C the right that A shall serve as B’svalet; and if A shall commit a breach it will be C who gets the damages measuredby the value of the promised service.118

115 AL Corbin, ‘Assignment of Contract Rights’ (1926) 74 University of Pennsylvania LawReview 207, 207.

116 Even so, this is really an abuse of the language of transfer since the obligee may stillbring legal proceedings against the obligor, such action being restrained only by theavailability of injunctive relief on the application of the assignee.

117 [2000] Ch 295, 321, aff’d [2000] Ch 291 (CA) [Don King].118 Above n 115, at 218 (emphasis in original).




Once the essence of what happens in an equitable assignment of a chose inaction is made plain, the difficulty with the statement in Rule 2 becomesmore obvious.119 And if that is the case, how can an anti-assignment clausehave any effect on this stripped-down and streamlined conception ofequitable assignment? Surely it cannot.

B. Anti-Assignment Clauses

Where does this bring us in relation to Lord Browne-Wilkinson’s applica-tion of the anti-assignment clause in Linden Gardens? With respect, it maybe that a quite different view would have been taken of the efficacy of theanti-assignment clause in clause 17(1) had the limited nature of anequitable assignment been appreciated. As noted above, an equitableassignment may operate as a contract between assignor and assignee. Thatcontract does not transfer (in the sense of a disposition) the benefit of thecontractual performance due from the obligor, but ‘transfers’ the assignor’sright to release the obligor from his or her contractual duties in conse-quence of the assignor’s promise to lend his or her name to the assignee forthe purposes of bringing an action on the subject matter of the assignment.As a corollary to the right to grant a release, the assignee would thereafteralso have the right to cause a claim to be brought against the obligorshould he or she default in the contractual performance.120 And certainly,where there is only a contractual promise to assign, because ‘equity lookson as done that which ought to be done’ to give effect to the assignor’sintentions, the courts of equity will recognise that an equitable interest inthe thing assigned (which is the right to bring legal proceedings on thechose in action, and not the chose in action itself) has been created and

119 Acknowledging Corbin’s views, Tolhurst recognises that this is a possible mode ofassignment: above n 25, at [6.75]. He describes this as being a case where ‘if an assignmentdoes not create privity of contract between the obligor and assignee, then arguably the effectof an assignment should be that the obligor continues to perform to the assignor but for theassignee; the performance is for the assignee as it is the assignee who owns the benefit of theright to performance and it is the assignee who can sue for breach of contract if the obligorfails to perform to the assignor. … [But t]he above possibility on its face takes no account ofnotice and dictates that this must be the effect of an assignment whether or not notice hasbeen given. Clearly that is not the law. There is no doubt that upon receipt of notice theobligor can generally obtain a discharge only by accounting to the assignee and that prior tonotice the obligor can obtain a discharge only by performing to the assignor’ (emphasis inoriginal). But, as has been explained elsewhere, this takes an overly reductionist view of howa contract might be discharged: see the discussion in Part II. C. (iii) and (iv), above.

120 Tolhurst accepts that the term ‘chose in action’ may refer to ‘a right enforceable byaction or to the right of action itself’: above n 25, at [2.02].

306 C H Tham



vested in the assignee.121 As some others have described it, what equitydoes is to create rights to rights. It does not transfer the underlying rightfrom one party to another.122

The scope of this equitable interest is necessarily very constrained if oneis also to honour the maxim that ‘equity follows the law’. The maxim that‘equity looks on as done that which ought to be done’ cannot bring theassignee into any form of direct relationship with the obligor to the chosein action, as to do this would be to replicate in equity something which thecommon law would not have permitted in the absence of novation orstatute. If an anti-assignment clause typically has no effect vis-à-vis an‘assignment’ by way of trust,123 that ought equally be true of an assign-ment in equity, since both, on this analysis, operate in a broadly similarfashion, although the specific constitutive intentions are distinct.124

Let us assume that the equitable assignment involves not a transfer (or atleast not the same conception of transfer as one might apply to a statutoryassignment or a novation) but the recognition of a new equitable interest inthe chose in action, arising out of the specific enforceability of theassignor’s promise to lend his or her name to the assignee for the purposeof bringing legal proceedings on the chose. If this is so, Lord Browne-Wilkinson’s view that an anti-assignment clause may invalidate an equita-ble assignment cannot apply to an equitable assignment by means of atrust, since in the latter case, it is obvious that no direct contractualrelationship is ever created between obligor and assignee. Yet because theprocesses by which equitable interests are created by the constitution of atrust or by means of an equitable assignment other than by constitution ofa trust are so similar, if a typically worded anti-assignment clause isineffective to prevent an assignment by way of trust (as was held to be thecase in Don King),125 the same should be true of an equitable assignment.

121 Once executed consideration for the assignment has been furnished to the assignee.The clearest statement of this principle may be found in Meagher, above n 46, at [6–050]. Butthe same point is repeated in Tolhurst, above n 25, at 334–5. As to whether executedconsideration had been furnished in the two appeals in Linden Gardens to the assignors inboth appeals, see n 38 above.

122 See R Chambers, An Introduction to Property Law in Australia, 2nd edn (ThomsonLawbook Co, Sydney, 2008) [13.90]; B McFarlane, The Structure of Property Law (HartPublishing, Oxford, 2008) 70–74. Indeed, to underscore profound distinction between thecommon law and equitable conceptions of ‘ownership,’ McFarlane proposes that we abandonattempts to describe property rights as being ‘equitable’ and recognise that in equity, one hassomething rather different: a ‘persistent right’ (ibid, at 70–71).

123 Don King, above n 117; approved by a majority of the Court of Appeal in BarbadosTrust Co Ltd v Bank of Zambia [2007] 1 Lloyd’s Rep 495 (CA) (Hooper LJ dissenting).

124 Tony Oakley would characterise the equitable interest in the equitable assignee wherethe assignment was for value to have arisen out of a constructive trust: see AJ Oakley,Constructive Trusts 3rd edn (London, Sweet & Maxwell, 1997) ch 8. Smith follows suit:above n 24, at [6.12].

125 Above n 117.




This, of course, inverts Andrew Tettenborn’s persuasive argumentsagainst the doctrinal soundness of Don King. Writing in response to theCourt of Appeal’s affirmation of Lightman J’s decision, Tettenbornobserved:

Let it be granted that [the assignor’s] rights under the management agreementswere held on trust for the partnership and hence were partnership property. Onthat assumption, just what rights did the partnership have in respect of them?Had it the right to control their exercise, or to insist on their being used in acertain way? Could it have disposed of its beneficial interest to anyone it thoughtfit? If the answer to these questions is Yes, then there was an equitableassignment of the agreements in all but name.126

But that is precisely the point—in equity, one may effect an assignmentboth by means of a trust or otherwise. If the anti-assignment clause doesnot prevent an equitable assignment by constitution of a trust, it shouldlikewise be unable to prevent an equitable assignment by means of therecognition of an equitable interest in the chose in action assigned.Although in the latter case equity is not recognising an equitable interest ofa trustee-beneficiary, it nevertheless recognises a slightly different form ofequitable interest, by reason of the specific enforceability of the assignor’scontractual promise to lend its name to the assignee for the purposes ofbringing legal proceedings on the chose in action assigned.

The Court of Appeal seems not to have shared Tettenborn’s views.127 Isuggest that the force of his criticisms may be blunted if we accept that theconcern with protecting the interests of an obligor may have beenoverstated. Tettenborn expands on the policy reasons for giving broadereffect to anti-assignment clauses as follows:

The object of allowing a contractor to stipulate that his obligations shall beunassignable is to allow him to obtain an assurance that he will have to dealwith his co-contractor and no-one else, and that no-one other than thatco-contractor will have the effective right to control the exercise of contractualrights against him.128

126 A Tettenborn, ‘Trusts and Unassignable Agreements–Again–Don King Productions vWarren’ [1999] Lloyd’s Maritime and Commercial Law Quarterly 352, 354. See also ch 11 ofthis book.

127 The majority of the Court of Appeal expressly adopted Lightman J’s analysis as to whyanti-assignment clauses do not typically prevent parties from constituting themselves astrustees of the benefit in a chose in action in the recent case of Barbados Trust Co Ltd v Bankof Zambia, above n 123. The position of the third member of the Court of Appeal in thiscase, Hooper LJ, is less easy to discern since it seems he both approbated and reprobatedLightman J’s analysis: ibid, at [139]. One therefore ought not to place too much weight onHooper LJ’s dissent. The point is discussed further in CH Tham, ‘What Assignments ofCauses of Action Are, and More—Offer-Hoar v Larkstore’ [2007] Lloyd’s Maritime andCommercial Law Quarterly 286, at text to fn 27.

128 A Tettenborn, ‘Trusts of Unassignable Agreements—Don King Productions v Warren’[1998] Lloyd’s Maritime and Commercial Law Quarterly 498, 499. For a concurring view,

308 C H Tham



However, were there to have been no assignment, even in a case where thecontractual promise is still wholly or partially executory, such assurance asthe contractor might have had in the manner in which a co-contractormight exercise its contractual rights is entirely non-binding. This isparticularly striking if we remember that an equitable assignment caneffect no change in the legal obligations undertaken by the contractor tothe co-contractor. Whether before or after assignment, the contractor’sobligations under the chose in action remain exactly the same, unless someform of variation for consideration has been effected or some form ofvariability has been built into the contractual terms.129 The question ofwhether the contractor is in breach of these obligations will still bemeasured against the terms of the original contract constituting the chosein action, and that is the precise question which the courts are to decideshould legal proceedings be brought on the chose.130

As with the constitution of a trust, an equitable assignment involves nodisposition, transfer or conveyance so as to bring the assignee and obligorinto direct contractual relations with each other. Like the case in which atrust is constituted over a chose in action, an equitable assignment of achose in action is an instance where equity is prepared to recognise thecreation and vesting of an equitable interest in the chose in action in theassignee. In effect, a new equitable interest in the chose in action is createdand vested in the assignee, rather than the chose in action being somehowtransferred from one party to another. Or, in other words, a new equitableinterest is ‘engrafted’ on to the chose in action. Therefore, even if clause17(1) in Linden Gardens had had the effect of negating transfers of thebenefit of the building contract without the consent of the obligor, it couldhave no effect on an equitable assignment of the benefit of the buildingcontract operating via something more akin to what happens in theconstitution of a trust of the chose.

C. Breaching an Anti-Assignment Clause: Ineffectiveness or Defeasibility?

The preceding discussion has attempted to demonstrate that an anti-assignment clause cannot prevent an assignor from effectively ‘transfer-ring’, in equity, his or her right to grant a release of the contractual chose

see G McMeel, The Construction of Contracts: Interpretation, Implication and Rectification(Oxford, Oxford University Press, 2007) [16.77].

129 Whether expressly, impliedly, or as a matter of construction. As argued above, the lastis the true ratio of Portland Cement.

130 Admittedly, there is one area where there is a discernible change as a result of anassignment, namely the promisor’s expectations as to the likelihood that legal proceedingsmight be brought against him or her. However, as the discussion below in the text following n137 shows, it is doubtful whether this is a legitimate interest deserving of judicial concern.




to an assignee of his or her choice. But we should not forget that ananti-assignment clause is, often, itself a contractual promise. The contentof that promise is a matter of construction, but at the very least it is apromise by an obligee who is subject to an anti-assignment clause that heor she will not effect an assignment to a third party. So, althoughanti-assignment clauses do not, on the analysis above, prevent equitableassignments from occurring, the breach of the promise not to effect anassignment cannot be without consequence. Roy Goode identified at leastone of these—where the breach of an anti-assignment clause constitutes abreach of contractual promise so severe that it gives the victim of thebreach the right to elect to discharge the entire contract. This is confirmedby the law relating to the forfeiture of leases following a breach of acovenant against assignment.

As Goode recognised, one possible interpretation of an anti-assignmentclause may be that it is a mere personal undertaking, ‘the breach of whichdoes not render the assignment ineffective against the debtor but merelyexposes him to a claim for damages for breach of contract’.131 In LindenGardens, Lord Browne-Wilkinson accepted Goode’s analysis that the legaleffect of an anti-assignment clause was, to begin with, a matter ofconstruction. He paraphrased Goode’s itemisation132 of the four possibleconstructions of an anti-assignment clause as follows:

(1) that the term does not invalidate a purported assignment by A to C but givesrise only to a claim by B against A for damages for breach of the prohibition; (2)that the term precludes or invalidates any assignment by A to C (so as to entitleB to pay the debt to A) but not so as to preclude A from agreeing, as betweenhimself and C, that he will account to C for what A receives from B … (3) thatA is precluded not only from effectively assigning the contractual rights to C, butalso from agreeing to account to C for the fruits of the contract when received byA from B; (4) that a purported assignment by A to C constitutes a repudiatorybreach of condition entitling B not merely to refuse to pay C but also to refuse topay A.133

Lord Browne-Wilkinson thought categories (1) and (4) to be unlikely,134

and certainly he did not arrive at his conclusion on the effect of theanti-assignment clause before him on the basis of discharge by breach. Butit is difficult to see just how clause 17(1) could ever have had the effect on

131 RM Goode, ‘Inalienable Rights?’ (1979) 42 MLR 553, 554. This is also the generalposition adopted in the UNIDROIT Principles of International Contracts 2004: see, eg, thecommentary to Art 9.1.3.

132 Ibid.133 Linden Gardens, above n 1, at 104.134 Ibid.

310 C H Tham



the assignee’s equitable interest in the chose in action assigned as LordBrowne-Wilkinson asserted it had, given the analysis put forward in Part IIand Part III.B above.

It may be necessary, however, to go further. First, recognising that thesource of an assignee’s equitable interest in a legal chose in action‘transferred’ to him or her by way of equitable assignment for value is theavailability of specific performance, it should cause little surprise if a courtshould decide against making such an order in circumstances where anassignee had notice that the legal chose in action was subject to ananti-assignment clause. In such cases, the equities between obligor andassignee would hardly be equal.

Second, if the anti-assignment clause functions as a contractual promisein its own right, the breach of that promise may well bring about certainconsequences which could discourage an obligor from breaching it, or aputative assignee from being involved in such a breach. Goode has alreadyadverted to the possibility that such a breach could, in appropriate cases,entitle an obligor to discharge the contract. In cases where such dischargeoccurs early enough, preceding the date on which the obligor’s contractualperformance becomes due, this would be a reason for a putative assignee tothink twice about his or her position. In such cases, even if the assignee hadbeen wholly ignorant of the anti-assignment clause, there would be nothingleft to specifically perform since the contractual obligations in the choseassigned would have been discharged by breach. What then of cases wheredischarge might conceivably occur after the contractual performance in thechose has fallen due? The simple discharge analysis will not help unless theobligor has been able to negotiate an anti-assignment clause whichfunctions not merely as promise but also as a condition subsequent.135 Ifso, the breach of the promise against assignment could well operate

135 On this, Lord Steyn’s explanation is helpful: ‘Traditionally, a distinction is madebetween conditions precedent and conditions subsequent. Given that one is dealing withcontingent as opposed to promissory conditions, one can for present purposes say that a factis a condition precedent to a contract for the creation of which it is necessary; and that a factis a condition subsequent to a contract that it extinguishes …’: Total Gas Marketing Ltd vArco British Ltd [1998] 2 Lloyd’s Rep 209 (HL) 221. That said, although the breach of ananti-assignment clause may well be a fact that would extinguish the chose in action of whichit is a part, that breach is also simultaneously the breach of a promise made by the assignor tothe obligor in that chose. As such, it is not open to the assignor-in-breach to rely on the factof its breach to assert that the chose in action has been entirely terminated: that option is leftfor the obligor to whom the promise of non-assignment is made. This is why the chose inaction does not automatically determine on the occurrence of the condition subsequent. Seealso n 152 below.




retrospectively,136 so as to deny legal effect to any contractual obligationsthat might have previously fallen due.137

The upshot, therefore, is that anti-assignment clauses may have the effectof deterring assignments in breach, but only as a matter of construction.No rule of law renders such assignments ineffective. If the policy reasonbehind such a rule of law is the desire to give effect to an obligor’s wishesnot to be brought into direct contractual relations with third parties suchas assignees of the obligee, these policy grounds are weak. That policypresumably derives from the obligor’s desire to minimise the risk of changein the manner and degree of supervision of the contractual performanceand of the likelihood that legal proceedings would be brought against himor her. Yet even in a simple two-party case, there is little an obligor can doto prevent a hitherto quiescent and cooperative counter-party from under-going the transformation from Dr Jekyll to Mr Hyde. Procedurally, theobligor may seek to convince the court that proceedings brought againsthim or her are unmeritorious. If so, such claims would be liable to bedismissed, presumably with an appropriate costs order. The obligor isexpected to perform to the standards set in the contract, and the frivolityof any decision to litigate on that contract will be measured against thatvery standard. Given the degree to which the rule as to precise performanceoperates so as to render invariable the obligations undertaken within thecontract, the fear that such contractually agreed standards and expecta-tions may wildly fluctuate is misplaced. It is unclear how assignment couldalter this.

It is true that the threat of legal proceedings brings about its own coststhat may be irrecoverable. Apart from the reality that the courts are

136 In Total Gas Marketing Ltd v Arco British Ltd, ibid, at 215 the House of Lordsrecognised that non-fulfilment of a condition subsequent could, as a matter of construction,have such an effect. Lord Slynn (with whom Lord Nolan, Lord Steyn and Lord Hope ofCraighead concurred) said: ‘If the provision in an agreement is of fundamental importancethen the result either of a failure to perform it (it if is promissory) or of the event nothappening or the act not being done (if it is a contingent condition or a condition precedent ora condition subsequent) may be that the contract either never comes into being or terminates.That may be so, whether the parties expressly say so or not. … To adapt the words of MrJustice Maugham in Re Sanderwell Park Colliery Co, [1929] 1 Ch 277 at p 282 “the veryexistence of the mutual obligations is dependent on the performance of the condition.” Forcompleteness I would substitute “performance or fulfilment of the condition” for “perform-ance of the condition”’ (emphasis in original). Lord Hutton’s speech makes much the samepoint, rejecting the submission that an unfulfilled contingent condition could only suspend theperformance of obligations under the contract and not terminate it: ibid, at 226. Further, ‘…in regard to contingent conditions, it is not necessary for parties when stipulating for acondition precedent or a condition subsequent to spell out the consequences of non-occurrence of the condition: these are prima facie inherent in the use of such terms …’, ibid,at 221 (Lord Steyn).

137 This may be the proper basis underlying the operation of an ‘anti-trusteeship’ clause(as envisaged in Don King, above n 117) or an ‘anti-equitable interest’ clause, absent notice ofsuch clauses.

312 C H Tham



reluctant to order indemnity costs on a routine basis, any threat of legalaction causes the putative defendant to incur time, labour and effort indealing with the claim. And if the proceedings were threatened by someoneother than the original counter-party with whom some form of commercialunderstanding had been struck, existing calculations could be upset. Butjust as with the actual bringing of proceedings, the risk of such proceedingsbeing brought and the calculations of those risks are, in the main, merespeculation. Unless there is a binding promise not to sue or to limit such aclaim, or if there is an operative estoppel, there would be nothing toprevent the original obligee from upsetting all prior expectations. Whyshould the assignee be put in any worse position?138 It is one thing for thelaw to impose burdens without an individual’s consent (as in tort, criminaland revenue law, for example). It would be an entirely different thing forthe law to allow individual private entities to make agreements to imposeburdens on strangers to that agreement.

Through appropriate drafting, one might be able to create a situationwhere the court will refuse to grant specific performance to an assigneebecause it had notice of an anti-assignment clause. And even without suchnotice, an appropriately worded anti-assignment clause might well beconstrued as a condition subsequent, so as to release the obligor from allcontractual obligations as yet unperformed, even if they are already due.139

All of this is consistent with established learning in relation to assignmentsof leases in breach of covenant.

Millett LJ reminded us in Hendry v Chartsearch Ltd that:

an assignment in breach of covenant is effective to vest the legal estate in theassignee: Old Grovebury Manor Farm v Seymour Plant Sales & Hire (No 2) …but the assignee takes a defeasible interest only which is liable to forfeiture forbreach of covenant.140

A landlord’s right to forfeit a lease may arise through (a) denial of thelandlord’s title, (b) breach of condition by the tenant, or (c) exercise of anexpress or implied forfeiture clause.141 So even in relation to leases, there isno blanket rule that the breach of a covenant against assignment withoutobtaining the landlord’s prior consent will render the assignment invalid:Old Grovebury Manor Farm v Seymour Plant Sales & Hire (No 2) tells us

138 Certainly if there were such promises, limits or estoppels, on the analysis herein suchrestrictions would similarly bind the assignee, for an assignee only obtains the right to directan obligee who has made an equitable assignment of the ‘chose in action’ in his or hercontract with the obligor to bring proceedings against the obligor in relation to the obligor’scontractual performance as owed to the assignor.

139 Though this might well raise issues pertaining to the availability of equitable reliefagainst forfeiture.

140 [1998] CLC 1382 (CA) 1393–4, [1998] EWCA Civ 1276, [45].141 AJ Oakley, Megarry’s Manual of the Law of Real Property, 8th edn (London, Sweet &

Maxwell, 2002) 354–5.




otherwise.142 Rather, the assignment is effective, though defeasible, andthat defeasibility is not automatic. Furthermore, ‘[e]ven if the landlord hasshown that he is treating the lease as forfeited, he may subsequentlyprevent himself from proceeding with the forfeiture if he waives the breachof covenant’.143 The parallel with the model proposed in this article seemstolerably close, and there is no obvious reason why assignments of a chosein action in contravention of an anti-assignment clause ought to be treatedany differently from assignments of leases.

This analogy was proposed but was rejected by the trial judge in StMartins.144 Judge Bowsher QC took the view that this analogy wasunhelpful:

The law does not come to the assistance of assignees of leases who have taken inbreach of covenant. On the contrary, the law assists the covenantee, even to theextent of holding that the covenant runs with the land and binds the assigneeseven though they are not mentioned in the covenant …

Leases are in a different case to bare contractual promises. A lease containsrights in contract and rights of estate. The assignment of a lease passes an estatewhereas an assignment of the benefit of a contract passes only a chose in actionwhich is only of benefit if it can be enforced in the courts.145

In this, the trial judge was supported by Lord Browne-Wilkinson whoobserved:

A lease is a hybrid, part contract, part property. So far as rights of alienation areconcerned a lease has been treated as a species of property. Historically the lawtreated interests in land, both freehold and leasehold, as being capable ofdisposition and looked askance at any attempt to render them inalienable. … Incontrast, the development of the law affecting the assignment of contractualrights was wholly different. It started from exactly the opposite position, viz.,contractual rights were personal and not assignable. … It is therefore notsurprising if the law applicable to assignment of contractual rights differs fromthat applicable to the assignment of leases.146

It is obvious that a lease is not entirely the same thing as a contractualchose in action. A lease is also a chattel real. But that difference is notrelevant to our present discussion, and it is hard to see why it should leadto a legal distinction. In recognising that a lease is a chattel real, the lawpermits burdens to run with the estate in a lease so as to outflank thecontractual rule against imposition of burdens on strangers to a contract.Although the doctrine of estates would also permit benefits to pass to those

142 [1979] 1 WLR 1397 (CA).143 Oakley, above n 141, at 355.144 Above n 1.145 Ibid, at 62.146 Linden Gardens, above n 1, at 108–109.

314 C H Tham



in privity of estate, there is nothing in that to preclude such benefits frompassing on the basis of contract. The rejection of the analogy with theposition in relation to leases by both the trial judge in St Martins and LordBrowne-Wilkinson in Linden Gardens is therefore regrettable.

Millett LJ’s judgment in Hendry v Chartsearch Ltd provides an alter-native explanation. There, he said that:

[i]n the case of a lease, the fact that an assignment in breach of covenant iseffective to vest the term in the assignee means that it is too late to seek consent;the breach of covenant is complete and the lease is liable to forfeiture. That isnot so in the case of the benefit of a contract. The assignment does not constitutea breach of contract and is without legal effect so far as the other party to thecontract is concerned. It is not too late for the assignor to ask for consent. Butthe contract requires the assignor to obtain the prior consent of the other party;retrospective consent, if given, may operate as a waiver, but cannot amount tothe consent require by the contract. The proper course if for the assignor to askfor consent to a new assignment and to wait until it is given or unreasonablyrefused to make it.147

There is little on the face of this statement, or in its context, to suggest thatMillett LJ was only putting forward what he thought to be a plausibleconstruction of the anti-assignment clause before him. The extract reads,perfectly naturally, as a generally applicable rationale that unless all thecontractual pre-conditions to assignment of a chose in action have beensatisfied, there is no assignment of the chose and no breach. Reiterating thepoint, Millett LJ said:

The assignment which was made without the prior written consent of thedefendants was effective as between assignor and assignee, but was ineffective asbetween the assignor and the defendants. The making of such an assignment didnot put the assignor in breach of contract, let alone in repudiatory breach; itsimply did not affect the [obligors’] legal position and could be disregarded bythem with impunity.148

But that would seem to introduce a circularity149: if the breach of thecontractual pre-condition as to non-assignability renders the assignmentineffective, then there is no breach; and if there is no breach, there ought tobe no impediment to the effectiveness of the assignment; yet if that were

147 Above n 140, at 1394.148 Ibid.149 This circularity has also been noted elsewhere: see GJ Tolhurst, ‘The Efficacy of

Contractual Provisions Prohibiting Assignment’ (2004) 26 Sydney Law Review 161, at text tofn 86. In that article, Tolhurst goes on to argue that the presence of an anti-assignment clause,being a promise not to assign, ‘… rendered the contractual right personal and not capable ofassignment’: ibid, at text following fn 88. But, as the discussion above suggests, so long as wekeep in mind the limited nature of the ‘transfer’ effected by an equitable assignment, there isno reason why personal obligations might not be assigned.




true, there would be a breach, and so on ad infinitum.150 The only way outof this circularity is to recognise that an anti-assignment clause does not,per se, invalidate an assignment in breach of the promise not to assign, butbreach of the promise not to assign may affect the chose in action whichhas been assigned. As suggested above,151 one possible outcome ariseswhere the anti-assignment clause is construed as a condition subsequent. Ifso, its breach will retrospectively nullify the contractual obligations con-tained in the chose, and that may be why the obligor, whose consent to theassignment has not been sought, is at liberty to ignore the assignment. Butthis does not invalidate the contract of assignment as between assignor andassignee. That assignment still stands, though as an assignment of subjectmatter which is defeasible at the option of the obligor.152 Therefore, MillettLJ’s assumption153 that assignments of choses in action and leases operatedifferently in this respect may bear re-consideration.

Returning to Linden Gardens, it should be noted that there was noexplicit consideration of these points, and therefore no finding that thebreach of the promise not to assign had given rise to a right in the buildingcontractors to discharge their obligations under the contract for breach ofcondition, much less a finding that such right of election had beenexercised. Nor was there any finding as to the application of any express orimplied contractual right to terminate the building contract. There wassimply a breach of the anti-assignment clause. The proposition in thisarticle is that non-compliance with an anti-assignment clause should haveno impact on the validity of the contract of assignment as between theassignor and the assignee, the very point accepted to be the case in Linden

150 One might try to finesse this, as Lord Browne-Wilkinson seems to have done, bysuggesting that it is the purported assignment without the prior consent of the obligor in theLinden Gardens appeals that amounts to a breach, above n 1, at 106. But can one really‘purport’ to do the impermissible? Is it really conceivable that contract law allows for theprescription of impossible intentions? Or might the rather less metaphysical model ofdefeasibility not be preferable? See also text to n 156 below.

151 See text to and following n 135 above.152 Helpfully, in Thompson v ASDA-MFI Group plc [1988] Ch 241, 266 Scott J clarified

the position set out in New Zealand Shipping v Société des Ateliers et Chantiers de France[1919] AC 1 (HL) and Cheall v Association of Professional Executive Clerical and ComputerStaff [1983] 2 AC 180 (HL) [Cheall] as follows: ‘In order to attract the principle that a partyis not entitled to rely on his own acts as fulfilling a condition subsequent and bringing acontract to an end, the act must be a breach of duty and, per Lord Diplock [in Cheall]: “theduty must be one that is owed to the other party under that contract; breach of a dutywhether contractual or non-contractual owed to a stranger to the contract does not suffice.”’The promise not to assign contained in an anti-assignment clause is plainly owed to theobligor to the original contract and it amounts, therefore, to an obligation owed to theobligor. The obligor may therefore elect to ignore such a breach, if that suits his or herpurposes. So on this analysis, it is not possible for the assignor-in-breach to unilaterally bringthe chose in action to an end by virtue of assignment of the chose in breach of theanti-assignment promise therein.

153 Above n 140, at 1394.

316 C H Tham



Gardens. The misstep appears to be the assumption that the breach of theanti-assignment clause would as a matter of law render ineffective assign-ments in breach thereof. Had the nature of equitable assignment, and themechanism by which breaches of anti-assignment clauses affect the con-tractual obligations assigned, been better appreciated, it might have beeneasier for the House of Lords in Linden Gardens to conclude that breachof clause 17(1) should not have invalidated the assignment of the benefitsof the building contracts in the two appeals. If so, there would have beenno black hole to fill.

IV. CONCLUSION

Linden Gardens appears to have been argued and decided upon twoassumptions: that equitable assignments have the effect of bringing acontractual obligor into direct contractual relations with third partyassignees, and that the rules of assignment permit contracting parties todefine limits as to when an assignment may be effective. But, as thepreceding analysis has sought to demonstrate, the basis for those fearsappears rather doubtful. That said, although this article takes the positionthat an anti-assignment clause does not render assignments in breachthereof ineffective as a matter of law, it recognises that it is possible thatbreach of an anti-assignment clause may, as a matter of construction, affectthe chose in action assigned, and that this might give an obligor someavenues of redress should its unwillingness to be subject to the whims of athird party assignee be thwarted. However, that outcome is premised ondefeasibility, as opposed to invalidity ab initio, of the assignment.

In opposition to this view of the limited effect of anti-assignment clauses,Tolhurst suggests that:

the doctrinal efficacy of a prohibition on assignment lies in the ability of theparties to a contract to mould those rights they bring into existence and robthose rights of what would otherwise be their inherent transferability. Thus,although it is no doubt correct to suggest that contract operates throughpromises, it arguably does not follow that any prohibition on assignment, nomatter how it is drafted, must amount to a mere promise not to assign. Thiswould no doubt be the case if a prohibition operated only as a matter ofcontract, but it is not the case where the prohibition is intended to characterisethe chose.154

With respect, this analysis downplays the significance of a breach of thecontractual promise contained in an anti-assignment clause, focusinginstead on the theory that an appropriately worded anti-assignment clause

154 Tolhurst, above n 25, at [6.83].




can and does make assignment in breach thereof impossible, because itqualifies the ‘transferability’ of that which is sought to be assigned. If thisview is applied to equitable assignment, it is questionable. As the analysisabove has sought to show, at law, legal choses in action are inherentlynon-transferable if, by that, we mean the ‘transfer’ of an obligor’scontractual obligations, such that it is to perform for the assignee insteadof the assignor. In the absence of novation, this type of ‘transfer’ can onlyoccur if the contract is construed so as to permit substitution or variationof the party to whom contractual performance is to be rendered. In somecircumstances, such a variation may occur through the process of accordand satisfaction, and in some others the rules of set-off may permit us toconclude that a discharge of the original contractual obligation has beeneffected. But in none of these circumstances can it be said that the originalobligation was discharged by performance.

Equitable assignment takes its place within this crowded space but doesnot operate in like manner. Rather, (a) if the assignor may be subject to anorder of specific performance in relation to the exercise of his or her rightto release the obligor from the contractual performance; then (b) theassignee may be said to have an equitable interest in the chose in action.That, loosely speaking, may well be described as a ‘transfer of the right togrant a release’.155

There may be a view that all of the arguments above to refute‘ineffectiveness’ of an assignment in favour of ‘defeasibility’ where theassignment has been made in breach of an anti-assignment clause likeclause 17(1) is merely semantic—that we can safely short-circuit theprocess by saying that there are instances where breach of a clause willrender an assignment ineffective. But in an echo of Millett LJ’s rationalisa-tion in Hendry v Chartsearch Ltd,156 that runs the risk of circularity.Surely the law can do better. Quite apart from the attractions of enhancingthe internal coherence of this corner of the law, there is advantage inreducing the legal risk posed by anti-assignment clauses to those whowould be prepared to purchase such assets. For with a reduction in thelegal risk, the discount applied to reflect that risk must, over time, reduceas well. Assignees face less risk, the assignors get more in exchange forwhat they are giving up, and obligors face no change whatsoever in thenature and extent of their legal obligations. What is there not to like aboutthat picture?

155 Together with its corollary, the right to bring legal proceedings (being the converse of arelease).

156 Above n 140.

318 C H Tham



13

Coming to Terms with The GreatPeace in Common Mistake

KELVIN F K LOW*

I. INTRODUCTION

TO AN OUTSIDER, the predominantly adverse reaction of aca-demic scholars to the decision of the English Court of Appeal inGreat Peace Shipping Ltd v Tsavliris Salvage must seem very odd.1

After all, the court sought to lay to rest the much-criticised decision ofDenning LJ in Solle v Butcher.2 It must surely be difficult to understandhow a decision to overrule a mistaken decision can itself be mistaken.Perhaps the reaction to The Great Peace demonstrates a mellowing ofacademic opinion over Denning LJ’s excesses in Solle v Butcher. Perhaps itis simply in the nature of academics to be contrarian, spurred on no doubtin recent years by an increasingly questionable pressure to publish morethan can humanly be read. While The Great Peace cannot be regarded asproviding complete and perfect justice, as the court itself readily admits,3 itis questionable if the best way forward is to reverse The Great Peace ratherthan build upon it.

The Great Peace and Solle v Butcher are, of course, concerned with aparticular species of mistake in contracting known as common mistake.Contractual mistake is of the ‘common’ variety when both parties to the

* Thanks are owed to Derek Davies and Francis Reynolds for their comments on amuch earlier draft of this article when it took the form of an article entitled ‘The Role ofEquity in Mistake’ that was presented at the ‘Exploring Contract Law’ symposium hostedby the University of Western Ontario in January 2008. I would also like to thank MindyChen-Wishart and Chee Ho Tham for helpfully pointing out that in its original form, thatarticle was far too broad in scope and overly ambitious to boot, something which seemsobvious in hindsight. This article focuses on the part of that earlier article which deals withcommon mistake in contract. My gratitude must also go out to Jason Neyers for theinvitation to participate in a very lively and enlightening symposium as well as mycolleague, Lusina Ho, for referring me to Jason.

1 [2003] QB 679 (CA) [The Great Peace].2 [1950] 1 KB 671 (CA).3 The Great Peace, above n 1, at [161].



contract share the same mistake. Although the label is sometimes also usedto refer to cases where both parties are contracting at cross-purposes,4 thisother variety of mistake is perhaps best referred to by its alias ‘mutualmistake’ to avoid confusion. To understand the controversy over theoverruling of Solle v Butcher by The Great Peace, it is necessary to studytwo other cases, amongst others, in some depth—the decision of the Houseof Lords in Bell v Lever Bros Ltd5 and the decision of Steyn J in AssociatedJapanese Bank (International) Ltd v Crédit du Nord SA.6 These four casesare the most important landmarks in the development of the law ofcommon mistake and in spite of doubts raised in The Great Peace, bothSolle v Butcher and Associated Japanese Bank continue to feature promi-nently in contract textbooks.7 However, rather than continue to look backto the past, I would suggest that The Great Peace provides us with theopportunity to look forward, a venture that I propose will prove morefruitful.

II. A ROUGH SKETCH OF THE LAW’S DEVELOPMENT

The acknowledged starting point for any study of the law of commonmistake is the decision of the House of Lords in Bell v Lever Bros Ltd,itself a notoriously difficult case to understand. Very briefly, the rather oddfacts of the case were as follows. The contract at the heart of the dispute isthat which Lever Bros Ltd entered into with Ernest Hyslop Bell and WalterEdward Snelling to terminate Lever Bros’ employment of the pair aschairman and vice-chairman respectively of Niger Company, Lever Bros’subsidiary. Lever Bros entered into this contract not knowing that theycould have terminated the pair’s employment contracts without paymentsince both had committed serious breaches of these contracts. To thesurprise of Lever Bros, Bell and Snelling testified that they had forgotten

4 As epitomised by Raffles v Wichelhaus (1864) 2 Hurl & C 906, 159 ER 375 (Ex Ct)wherein the parties purported to contract to buy and sell 125 bales of Indian cotton thatwould arrive in Liverpool on the ship Peerless from Bombay. It transpired that two shipsnamed Peerless were to arrive in Liverpool from Bombay, one departing in October and theother departing in December. The buyer thought that the contract referred to the formerwhereas the seller thought that the contract referred to the latter. In light of these facts, thecourt concluded that a contract had not arisen.

5 [1932] AC 161 (HL).6 [1989] 1 WLR 255 (QB) [Associated Japanese Bank].7 See, eg, HG Beale, Chitty on Contracts, Volume I: General Principles, 29th edn

(London, Sweet & Maxwell, 2004) 5–026–5–047; E Peel, Treitel: The Law of Contract, 12thedn (London, Sweet & Maxwell, 2007) 328–30; SA Smith, Atiyah’s Introduction to the Lawof Contract, 6th edn (Oxford, Clarendon Press, 2005) 180–81. See also J Cartwright,Misrepresentation, Mistake and Non-Disclosure, 2nd edn (London, Sweet & Maxwell, 2007)15.27–15.29.

320 Kelvin F K Low



about their breaches of duty when they entered into the terminationagreements, setting the stage for an argument on common mistake.8

It is a perilous task to attempt to derive a ratio from Bell v Lever BrosLtd but two points may be stated with some certainty. First, it is not everycommon mistake which will affect the validity of a contract but onlyextreme cases. Although various epithets can and have been used todescribe the type of common mistake required to be established, it is reallydifficult to do better than ‘fundamental’.9 It is debateable what ‘fundamen-tal’ means but the majority of the House of Lords in Bell v Lever Bros Ltdset an exceedingly high bar for passing the test given the unquestionableseriousness of the mistake in the case. The majority therefore appears tosuggest that the scope of the nascent doctrine of common mistake is anexceedingly narrow one. Indeed, it paints so narrow a picture as to callupon itself criticism that it has been wrongly decided on its facts.10

Secondly, the effect of common mistake at common law is to nullifyconsent and render the contract void ab initio. To understand this aspect ofthe decision, it is necessary to appreciate the background in which the casewas decided. Early cases of mistake, it has been observed, tended not toreason from a doctrine of mistake, but either from an implication of acondition precedent as to the absence of the mistake11 or on the basis oftotal failure of consideration.12 ‘This was an era when the parties’agreement was everything.’13 Although the judges in Bell v Lever Bros Ltddo refer to common mistake simpliciter, traces of the earlier ‘implied term’theory of common mistake can still be found in the case.14 Under theinfluence of the ‘implied term’ theory of common mistake, the conclusionthat any operative common mistake must render the contract void ab initioseems inevitable. If the doctrine of common mistake is to be alternativelyanalysed in terms of implied conditions precedent, then the result of the

8 C MacMillan, ‘How Temptation Led to Mistake: An Explanation of Bell v Lever BrosLtd’ (2003) 119 LQR 625, 645–6.

9 JC Smith, ‘Contracts—Mistake, Frustration and Implied Terms’ (1994) 110 LQR 400,401.

10 Ibid, at 414–15.11 AWB Simpson, ‘Innovation in Nineteenth Century Contract Law’ (1975) 91 LQR 247,

268; Barr v Gibson (1838) 3 M & W 390, 150 ER 1196; Couturier v Hastie (1856) 5 HLC673, 10 ER 1065; Pritchard v Merchant’s and Tradesman’s Mutual Life-Assurance Society(1858) 3 CB(NS) 622, 140 ER 885.

12 Strickland v Turner (1852) 7 Exch 208, 155 ER 919.13 J Cartwright, ‘The Rise and Fall of Mistake in the English Law of Contract’ in R

Sefton-Green (ed), Mistake, Fraud and Duties to Inform in European Contract Law(Cambridge, Cambridge University Press, 2005) 65, 73. Not only were common mistake andfrustration attributed to the parties’ presumed agreement, so too were the proper law of thecontract (Lloyd v Guibert (1865) LR 1 QB 115, 120–21) and implied terms (The Moorcock(1889) 14 PD 64 (CA) 68).

14 Bell v Lever Bros Ltd, above n 5, at 206 (Lord Warrington with whom ViscountHailsham concurred) and at 224–7 (Lord Atkin with whom Lord Blanesburgh concurred).

The Great Peace and Common Mistake 321



doctrine must surely be to render the contract void ab initio since that isthe basis upon which conditions precedent operate.

The narrowness of the common law doctrine of common mistake wasseized upon by Denning LJ in Solle v Butcher to lay down a sister doctrineof common mistake in equity. In Solle v Butcher, the plaintiff, GodfreyFrank Solle, leased a flat from the defendant, Charles Butcher. The flat hadpreviously been let at a rent of £140 per annum. Owing to damage by amine in the war, the flat been repaired and substantially altered and bothparties mistakenly thought that this freed the premises from rent control.They thus agreed to a rent of £250 per annum. After the relationshipbetween the parties soured, Solle sought a declaration that the rent for theflat was £140 per annum and to recover his overpayments. Butcherresponded by appealing to the court to set aside the lease for commonmistake. The majority of the Court of Appeal expressed the view that thecontract was voidable in equity for common mistake but that the contractought to be rescinded upon terms being imposed in equity. This wasbecause it transpired that Butcher could probably have charged close to£250 per annum for the flat if he only had complied with the statutoryprocedures which would have allowed him to make additions to thestandard rent in respect of the improvements and structural alterations tothe flat. Solle v Butcher marks the beginnings of the dissociation of thedoctrine of common mistake from that of implied conditions precedent.According to Denning LJ, apart from cases where such a conditionprecedent can be implied into the contract, common mistake, howeverfundamental, did not render a contract void.15 Therefore, the reality is thatthere is no separate and independent doctrine of common mistake atcommon law at all. Instead, common mistake as a separate doctrine, if itoperated at all, operated purely in equity where the parties laboured undera common fundamental mistake, provided the party seeking to set asidethe contract was not himself or herself at fault.16 Where it operated, itrendered the contract voidable and not void and the courts retained adiscretion to impose terms in awarding rescission.

Denning LJ’s judgment in Solle v Butcher has been criticised for anumber of different reasons. First, it has been suggested that Denning LJ’sinterpretation of the common law doctrine in Bell v Lever Bros Ltd wasoverly restrictive,17 though it is arguable that this criticism is purelysemantic since some of their Lordships in Bell v Lever Bros Ltd did appearto refer to the doctrine of common mistake interchangeably with that of

15 Above n 2, at 691–3.16 Ibid, at 693.17 See Associated Japanese Bank, above n 6, at 267 where this suggestion is made.

322 Kelvin F K Low



implied conditions precedent.18 Perhaps more importantly, it has beenconvincingly established that the cases from which Denning LJ derived theequitable jurisdiction in fact provided little authority for such a view.19 Hisappeal to cases of misrepresentation and rectification for unilateral mistakedo not support his view of the equitable jurisdiction in cases of commonmistake since those situations clearly raise very different concerns.20

The apparent tension between Bell v Lever Bros Ltd and Solle v Butcherwas addressed by Steyn J in Associated Japanese Bank in the third of ourquartet of leading English cases. Here, Bennett contracted to sell andleaseback from the Associated Japanese Bank (International) Ltd fourmachines which did not exist. Crédit du Nord SA guaranteed Bennett’sperformance under the leaseback agreement. After making a single pay-ment under the leaseback agreement, Bennett defaulted and the AssociatedJapanese Bank sued Crédit du Nord on the guarantee. The significance ofAssociated Japanese Bank is that it clearly dissociates the doctrine ofcommon mistake from the implied condition precedent theory both at lawand in equity. According to Steyn J:

Logically, before one can turn to the rules as to mistake, whether at common lawor in equity, one must first determine whether the contract itself, by express orimplied condition precedent or otherwise, provides who bears the risk of therelevant mistake.21 … Only if the contract is silent on the point, is there scopefor invoking mistake … Where common law mistake has been pleaded, the courtmust first consider this plea. If the contract is held to be void, no question ofmistake in equity arises. But, if the contract is held to be valid, a plea of mistakein equity may still have to be considered.22

Having thus differentiated for the first time common mistake at commonlaw from the implied term theory, Steyn J accommodated Solle v Butcherby suggesting that the equitable jurisdiction to provide relief from commonmistake was wider than that of the common law. This three stage test thuspurports to reconcile Solle v Butcher with Bell v Lever Bros Ltd. AlthoughAssociated Japanese Bank’s claim on the guarantee fell at the first hurdle,Steyn J suggested by way of dicta that even if that had not been the case,the contract of guarantee would in any event have been void at law forcommon mistake. In this respect, it rejects a narrow interpretation of Bell vLever Bros Ltd which would restrict the common law doctrine to cases

18 Above n 5, at 206 (Lord Warrington with whom Viscount Hailsham concurred) and at224–7 (Lord Atkin with whom Lord Blanesburgh concurred).

19 CJ Slade, ‘The Myth of Mistake in the English Law of Contract’ (1954) 70 LQR 385.20 J Cartwright, ‘Solle v Butcher and the Doctrine of Mistake in Contract’ (1987) 103

LQR 594.21 Such risk allocation may also be the result of the rules of general law applicable to the

contract: William Sindall Plc v Cambridgeshire County Council [1994] 1 WLR 1016 (CA)1035 (Hoffmann LJ).

22 Associated Japanese Bank, above n 6, at 268.




where there is a mistake as to the existence of the subject matter of thecontract. On its facts, even though the machines did not exist, themachines were not themselves the subject matter of the contract ofguarantee. Rather, the leaseback contract was the subject matter of thecontract of guarantee and while the leaseback agreement was voidable forfraud, it was not void. So analysed, however, the similarities betweenAssociated Japanese Bank and Bell v Lever Bros Ltd become obvious. ‘Inboth, the subject-matter of the contract alleged to be affected by themistake was an earlier contract; and in both that earlier contract was liableto be rescinded by one of the parties to the subsequent contract.’23

Although there are differences between the two cases, they are notsatisfactory grounds for distinguishing them.24 Instead, Bell v Lever BrosLtd is distinguished as ‘a quite exceptional case’ featuring peculiar facts.25

Indeed, it would not be unfair to suggest that Bell v Lever Bros Ltd is farless influential than a decision of the House of Lords would ordinarily beand some may even suggest that Steyn J had merely paid lip service to thecase.26

For almost 15 years, Associated Japanese Bank stood as the leadingauthority on the analytical approach to the subject of common mistake.Then the Great Peace sailed and an ill-advised piece of litigation changedthe landscape of the subject. In The Great Peace, a ship, the CapeProvidence, suffered serious structural damage in the Indian Ocean. Whenit learnt of this, the defendant, Tsavliris (International) Ltd, offered itssalvage services, an offer which was accepted by the owners of the CapeProvidence. As the tug allotted to the task was still a good five or six daysaway from the Cape Providence, Tsavliris approached a firm of Londonbrokers who negotiated with the claimants, Great Peace Shipping Ltd, witha view to chartering the Great Peace to escort the Cape Providence untilthe tug arrived. A charter was concluded between the parties for aminimum of five days in the belief that the Great Peace was some 35 milesdistant from, and the nearest vessel to, the Cape Providence. It transpiredthat the Great Peace was actually 410 miles away from the CapeProvidence but Tsavliris did not immediately cancel the charter of theGreat Peace upon learning of their mistake. Instead, it first sought to locate

23 GH Treitel, ‘Mistake in Contract’ (1998) 104 LQR 501, 504.24 Treitel rejects, eg, the differences in the grounds of rescission and the fact that the earlier

contract in Bell v Lever Bros Ltd involved the same parties whereas in Associated JapaneseBank, the earlier contract involved one party to the contract allegedly affected by mistake anda third party (ibid, at 504–505).

25 Treitel, above n 23, at 505–507; See also MacMillan, above n 8.26 Treitel suggests that the difference in results in the two cases, whilst not exposing a

direct conflict between them, demonstrates a conflict of policies (ibid, at 507). Whereas Bell vLever Bros Ltd emphasised the need to respect the sanctity of contract, Associated JapaneseBank emphasised and gave effect to the reasonable expectations of honest men.

324 Kelvin F K Low



a nearer vessel. Fortuitously, another vessel chartered by the same charter-ers as the Cape Providence, the Nordfarer, happened to pass the CapeProvidence, and Tsavliris entered into new arrangements for the Nordfarerto serve as escort for the Cape Providence. Tsavliris then sought to cancelthe charter for the Great Peace some two hours after it had altered courseto assist the Cape Providence. Whereas the agents of the Great Peaceoffered to persuade her owners to accept two days hire, Tsavliris refused toaccept the conciliatory gesture. When they were sued upon the charter,Tsavliris argued that the entire contract was void for common mistake.

Considering their behaviour upon learning of the mistake, it shouldcome as no surprise that their defence to the action failed, both at firstinstance before Toulson J27 and on appeal to the Court of Appeal.Remarkably though, considering the clear facts of the case and the rules onprecedent,28 both Toulson J and the Court of Appeal sought to restate theEnglish law of common mistake by overruling Solle v Butcher andconsigning the equitable jurisdiction to intervene to the history books. Indoing so, the Court of Appeal confirmed that the doctrine of commonmistake at common law was not the result of some implied conditionprecedent but was a separate rule of law.29 Whilst criticism on the basis offailure to adhere to the rules on precedent is not unexpected, it issomewhat odd, considering the pedigree of Solle v Butcher, to find thatThe Great Peace has received a generally frosty reception from bothacademics30 and judges from other Commonwealth jurisdictions.31 Itappears that, despite initial hostility, Solle v Butcher has in recent yearscome to be regarded more favourably. Only two years prior to The GreatPeace, Sir Christopher Staunton had expressed the view that Solle vButcher ‘can on occasion be the passport to a just result’.32

27 Great Peace Shipping Ltd v Tsavliris Salvage (International) Ltd (2001) 151 NLJ 1696(QBD) [The Great Peace (QB)].

28 G McMeel, ‘“Equitable” Mistake Repudiated: The Demise of Solle v Butcher?’ [2002]Lloyd’s Maritime and Commercial Law Quarterly 449; SB Midwinter, ‘The Great Peace andPrecedent’ (2003) 119 LQR 180; D Sheehan, ‘Vitiation of Contracts for Mistake andMisrepresentation of Law’ (2003) 11 Restitution Law Review 26, 33.

29 The Great Peace, above n 1, at [73].30 FMB Reynolds, ‘Reconsider the Contract Textbooks’ (2003) 119 LQR 177; A Phang,

‘Controversy in Common Mistake’ [2003] Conveyancer and Property Lawyer 247; JDMcCamus, ‘Mistaken Assumptions in Equity: Sound Doctrine or Chimera?’ (2004) 40Canadian Business Law Journal 46.

31 Chwee Kin Keong v Digilandmall.com Pte Ltd [2005] SGCA 2, [74]; Miller Paving Ltdv B Gottardo Construction Ltd (2007) 31 BLR (4th) 33 (Ont CA). On the former, see TMYeo, ‘Great Peace: A Distant Disturbance’ (2005) 121 LQR 393 and PW Lee, ‘UnilateralMistake in Law and Equity—Solle v Butcher Reinstated’ (2006) 22 Journal of Contract Law81.

32 West Sussex Properties Ltd v Chichester DC [2000] NPC 74 (CA) [42] [West SussexProperties].




III. CRITICISMS AND ANALYSIS

To the extent that the overruling of Solle v Butcher is justified by thesuggestion that equitable intervention undermines the policy of the com-mon law in maintaining commercial certainty,33 the decision of The GreatPeace cannot be sustained. It is ‘elementary’ that there must be a point intime at which a doctrine would have no precedent simply because it wouldbe the first precedent to enunciate the principle concerned.34 It is also notuncommon for equity to ‘undermine’ the common law by preferringdifferent policies.35 There is an admitted need in this area of the law tobalance between the need to respect the sanctity of contract and the needto give effect to the reasonable expectations of honest men36 and thesuggestion that commercial certainty is of paramount concern cannot beseriously entertained. The Court of Appeal in The Great Peace did itself nofavours by conceding that it would welcome greater flexibility than thecommon law permitted.37 In suggesting that such development was beyondthe common law and could only be provided through Parliamentaryintervention, they failed to acknowledge that such statutory reform ishardly a priority for law reformers.38 The Court of Appeal also suggeststhat, in resting ‘fundamental’ in equity upon notions of fairness,39 the testin equity is far too uncertain,40 but there is surely an element here of thecommon law pot calling the equitable kettle black.

The strongest case in favour of the reform undertaken by the Court ofAppeal lies in the difficulty in disentangling the uncannily similar jurisdic-tions at common law and in equity, particularly since most cases will beaddressed by the quite separate question of express or implied contractualallocations of risk. How does one distinguish between ‘essentially different’at law from ‘fundamental’ in equity?41 Neither test is completely precise,though imprecision in this context is neither fatal nor avoidable. Juxta-posed together, however, their interaction creates further, and unnecessary,uncertainty since both tests, occasionally identically worded, seek tocapture the same idea. At some point when the parties’ agreement runs out

33 The Great Peace, above n 1, at [156].34 Phang, above n 30, at 251–2.35 For example, until the enactment of s 25(7) of the Judicature Act 1873 stipulations

specifying the time of performance were generally regarded as ‘of the essence’ at common law,whereas the reverse was true in equity (Parkin v Thorold (1852) 16 Beav 59, 51 ER 698).

36 Treitel, above n 23, at 507.37 The Great Peace, above n 1, at [161].38 Reynolds, above n 30, at 179.39 As Lord Denning MR envisaged in Magee v Pennine Insurance Co Ltd [1969] 2 QB 507

(CA) 514–15.40 The Great Peace, above n 1, at [138]. See also the criticism by Toulson J: ‘Bluntly, the

difficulty about this form of the doctrine is that it puts palm tree justice in place of partyautonomy.’ The Great Peace (QB), above n 27, at [120].

41 The Great Peace, above n 1, at [131], [154].

326 Kelvin F K Low



(that is when the case cannot be decided on the basis of contractualallocation of risk, either express or implied), there may be occasion for thecourts to relieve the parties of their bargain on the basis of commonmistake. To balance between the need to protect the sanctity of contractand the reasonable expectations of honest men, such intervention will berare and the mistake must be sufficiently serious before the courts willintervene. Whether the basis for intervention is described as ‘essentiallydifferent’ or ‘fundamental’, both epithets encapsulate the same idea—striking the appropriate balance. To suggest that the balance is differentlystruck at law and in equity, especially given the necessarily inexact terms ofboth tests, invites unnecessary uncertainty.

Despite the very real difficulty raised as to how a double test system withidentical concerns and similar inexactitude can realistically be applied,critics of The Great Peace suggest that other concerns ought perhaps totake precedent. These number three. First, there is concern that thecommon law jurisdiction as expressed in Bell v Lever Bros Ltd andreformulated in The Great Peace is simply too narrow42 and equity’sintervention is perhaps necessary to supplement the law. Secondly, con-cerns have been raised about the potential for third party rights to beimplicated by the effects of common mistake at law as the doctrine causesthe contract to be avoided ab initio rather than be rendered merelyvoidable.43 Finally, and perhaps most critically, the loss of remedialflexibility in equity is regarded as lamentable.44

A. The Limits of the Common Law

To suggest that the result in Bell v Lever Bros Ltd casts the doctrine ofcommon mistake at common law in too narrow a light is not to make aparticularly profound statement. It was convincingly demonstrated morethan 10 years ago that the facts of the case could easily have been treatedas a case of either res sua or res extincta (depending on whether oneadopted the perspective of Lever Bros or that of Bell and Snelling) and oneither view, the contract should have been void at law.45 A more recent,and very telling, historical analysis of the case has cast further doubts as tothe correctness of the case.46 Although there are suggestions in Bell v LeverBros Ltd that it was not clear, given Lever Bros anxiousness to terminatethe two agreements as a result of the merger of the Niger with its rival, thatLever Bros would not have entered into the disputed contract to terminate

42 McCamus, above n 30, at 77–8.43 Phang, above n 30, at 252–3; McCamus, ibid, at 79.44 Reynolds, above n 30, at 179; Phang, ibid, at 252–3; McCamus, above n 30, at 80–81.45 Smith, above n 9, at 414–15.46 MacMillan, above n 8.




the employment of Bell and Snelling even if they had known of thevoidability of their employment contracts,47 this seems altogether unrealwhen the facts are properly appreciated. It has been suggested, quiterightly, that by the time the case reached the House of Lords, the case hadbecame detached from its facts.48 Bell and Snelling’s egregious breacheshad transformed into foolish mistakes which caused no harm to LeverBros. Instead of being recognised as managers of a huge corporate concern,Bell and Snelling are cast as mere ‘servants’.49 The focus turned from theirbreaches to their exceptional service to the Niger Company. The suitinstituted by Lever Bros came to be viewed, not as an action concernedwith commercial honesty,50 but as an action by a greedy corporation benton trying to save money.51 It is also significant that the case was decided adecade before directors’ duties came to be authoritatively expounded inRegal (Hastings) Ltd v Gulliver.52 Viewed through more modern lenses,many of these concerns seem out of place. It is orthodox fiduciary law thatbreaches are not excused even if they resulted in no harm to thebeneficiary. Indeed, they are not excused even if the breaches themselvesresulted in the beneficiary making a tidy profit.53 How then can it berelevant that the fiduciary had, apart from the breach, performed exem-plary service? The suggestion that Bell and Snelling were mere ‘servants’would today be regarded as so absurd that any lawyer making such asuggestion would likely be laughed out of court. As a result, Bell v LeverBros Ltd is difficult to reconcile with more recently decided cases on thesubject of fiduciary law such as Item Software (UK) Ltd v Fassihi.54 Bell vLever Bros Ltd is very much a decision coloured by concerns which aretoday irrelevant. However, most modern cases do little more than pay lip

47 Bell v Lever Bros Ltd, above n 5, at 236 (Lord Thankerton).48 MacMillan, above n 8, at 650.49 See the various divergent examples constructed by Lord Atkin in his judgment to

support his conclusion: Bell v Lever Bros Ltd, above n 5, at 228.50 As it had been at trial: see Bell v Lever Bros Ltd, above n 5, at 625 in which the closing

submission of Lever Bros Ltd’s counsel, Stuart Bevan KC, at trial is quoted: ‘Do you think thisaction is brought for £30,000 or £20,000? … This action … was essential to bring, because ifthis sort of thing goes on and is permitted to go on, the whole fabric of commercial honestyand the whole structure of business becomes rotten.’

51 Bell v Lever Bros Ltd, above n 5, at 213 (Lord Atkin). This suggestion does not accordwith the real motivation for prosecuting the action. According to MacMillan, above n 8, at640: ‘It was unlikely to have been brought to recover the severance sums for the sake of themoney alone. The existing Lever Brothers legal files disclose no attempts to settle the action.Lever Brothers budgeted £20,000 to prosecute the action. They persisted when the costs hadrisen and they estimated that the cost of the action would be £30,000 of which they mightrecover £15,000. In the end, they spent in excess of £40,000. This was a suit brought uponprinciple and not for profit.’

52 [1967] 2 AC 134 (HL).53 Boardman v Phipps [1967] 2 AC 46 (HL).54 [2004] EWCA Civ 1244. Note, however, the criticisms of the case by A Berg, ‘Fiduciary

Duties: A Director’s Duty to Disclose his Own Misconduct’ (2005) 121 LQR 213.

328 Kelvin F K Low



service to Bell v Lever Bros Ltd in any event.55 It may be best for Bell vLever Bros Ltd to be laid to rest but the likelihood of litigation overcommon mistake recurring, much less reaching the House of Lords, seemsremote at best. Lip service seems to be the tidiest solution in the interim.Certainly the recasting of the test for common mistake by the Court ofAppeal in The Great Peace in terms equivalent to those applicable to bothfrustration and fundamental breach suggest that Bell v Lever Bros Ltd willcast nothing more than a pale shadow in the future.

However, this linking of the test for common mistake to that offrustration and fundamental breach has been criticised as a regression. It issuggested that it ‘more or less returns the analysis to one of being requiredto determine whether the subject matter of the contract is no longer inexistence, in some sense, and therefore cannot be delivered’.56 Such anarrow test, it is said, will drive the courts to manipulate the concept of‘the subject matter of the agreement’ so as to produce the desired result.57

It is not self-evident that the Court of Appeal actually does this. It hasnever been clear that the oft-cited test at common law of ‘essentiallydifferent’ was meant to serve as anything more than a metaphor for a testthat cannot be precisely stated and is intended to illustrate the need tobalance between competing policy interests. In citing Diplock LJ’s pioneer-ing judgment in Hongkong Fir Shipping Co Ltd v Kawasaki Kisen KaishaLtd in which his Lordship opines that ‘[t]he test … has been stated in anumber of metaphors all of which … amount to the same thing: does theoccurrence of the event deprive [a party] of substantially the whole benefitwhich it was the intention of the parties as expressed in the contract thathe should obtain as the consideration for [performance]?’58 it is not evidentthat the Court of Appeal was returning common mistake to the Jurassicage where it was believed that only res sua or res extincta would render thecontract void. If anything, this reformulation helpfully frees the law ofcommon mistake from the vice-like grip of Bell v Lever Bros Ltd. Thesuggestion that the link to frustration and therefore impossibility calls forsuch an interpretation of The Great Peace takes the idea of impossibility infrustration far too literally. ‘Impossible’, it has been said, is ‘something of arelative term’.59 It cannot mean literal impossibility, else Taylor v Cald-well60 should have been decided differently.61 However, the English courtshave generally resisted the temptation to transition to the more lenient

55 See, eg, Associated Japanese Bank, above n 6.56 McCamus, above n 30, at 78.57 Ibid.58 [1962] 2 QB 26 (CA) 66.59 Peel, above n 7, at 940.60 (1863) 3 B & S 826, 122 ER 309.61 Peel, above n 7, at 940, citing LL Fuller and MA Eisenberg, Basic Contract Law, 3rd

edn (St Paul, West Publishing, 1972) 801.




‘impracticable’ because, the test being vague enough as it is, it is necessaryto drive home the message that sanctity of contract is a very real concern tothe courts in balancing the competing policy interests. It is certainlyquestionable whether an open-textured rule drawn in terms of ‘fundamen-tal’ or ‘basic’62 is either excluded by the terms of The Great Peace orwould provide greater guidance.

The suggestion that American law is superior in this respect is question-able.63 According to § 152 of the Restatement (Second) of the Law ofContracts, the test for common mistake is as follows:

(1) Where a mistake of both parties at the time a contract was made as to a basicassumption on which the contract was made has a material effect on the agreedexchange of performances, the contract is voidable by the adversely affectedparty unless he bears the risk of the mistake under the rule stated in § 154.64

An English or Canadian lawyer would be alarmed by the label ‘material’but it is stressed that ‘material’ in American means something quitedifferent than ‘material’ in English.65 This begs the question why it issuperior? Both terms—‘material’ and ‘impossible’—are relative and equallycapable of manipulation. Both terms require further explanation as to thecompeting policies they seek to balance. In American law, the Commentaryto § 152 provides:

The mere fact that both parties are mistaken … does not, of itself, afford areason for avoidance of the contract by the adversely affected party. Relief isonly appropriate in situations where a mistake of both parties has such amaterial affect on the agreed exchange of performances as to upset the very basisfor the contract … It is not enough for [a party] to prove that he would not havemade the contract had it not been for the mistake. He must show that theresulting imbalance in the agreed exchange is so severe that he can not fairly berequired to carry it out.66

It is not immediately evident that the American test or descriptionencapsulates a markedly different idea or captures the same idea moreeffectively than Diplock LJ’s test in Hongkong Fir. Surely then the choiceof labels—between ‘material’ and ‘impossible’—merely serves to highlightthe message a particular legal system wishes to send to potential litigants asto which way the system leans in the application of a test as vague andopen-textured as that which applies in the context of common mistake,frustration and fundamental breach. Are the courts suggesting that theylean in favour of sanctity of contract, thereby discouraging litigation; or

62 McCamus, above n 30, at 78.63 Ibid, at 63–66, 76–79.64 Restatement (Second) of the Law of Contracts (1981) § 152.65 McCamus, above n 30, at 63–4.66 Restatement (Second) of the Law of Contracts, above n 64, at 386.

330 Kelvin F K Low



that they lean in favour of providing relief from the potentially very seriousconsequences of mistake, thereby encouraging litigation?67 It is not self-evident that either view is preferable. Nor, given the rarity of cases ofcommon mistake, does it seem practical for the courts to agonise over soslight a difference or academics to spill so much ink over it. Everyone willhave his or her own preference as to a choice between the two labels butthere are surely more important issues to mull over than the label to thealready flexible test for common mistake. After all, the vast majority ofcases will be decided on the basis of express or implied contractualallocations of risk.68

B. The Protection of Third Parties

The protection of innocent third parties is a concern which, at first blush,appears worthy of concern, but on careful examination proves to be muchless deserving. The idea that innocent third parties need protecting wasfirst pursued by Denning LJ in Solle v Butcher.69 It is said that if thecontract is void at common law, no property will pass to the originalpurchaser. Nemo dat quod non habet, the original purchaser will thus failto pass title on to the innocent third party purchaser. If the effect is insteadequitable rescission, then relief can be withheld where the interests of abona fide third party purchaser intervenes. A number of responses imme-diately spring to mind. First, the preservation of the equitable doctrine, inand of itself, does nothing to protect third parties. If, as envisaged prior toThe Great Peace, equity intervenes only where the common law does not,any third parties that would be prejudiced by the common law doctrinewould still remain prejudiced notwithstanding the equitable jurisdiction. Itis only if equity is allowed to swallow up the common law that one may

67 In this context, litigation should not be regarded as carrying any particular negativeconnotations. Save in very clear-cut cases, such as res sua or res extincta, parties are unlikelyto agree as to whether or not the mistake is sufficiently serious as to lead to the discharge ofthe contract. Discouraging litigation may lead to some mistaken parties suffering the adverseconsequences of the mistake even though they have a legitimate claim to relief because theyare reluctant to seek relief.

68 As Hoffmann LJ reminded us in William Sindall, above n 21, at 1035: ‘When [Steyn J inAssociated Japanese Bank] speaks of the contract allocating risk “by express or impliedcondition precedent or otherwise” I think he includes rules of general law applicable to thecontract and which, for example, provide that, in the absence of express warranty, the law iscaveat emptor.’ The already short list of cases on common mistake are then whittled downfurther when his Lordship observed that ‘neither in Grist v Bailey [1967] Ch 532 nor inLaurence v Lexcourt Holdings Ltd [1978] 1 WLR 1128 did the judges who decided thosecases at first instance advert to the question of contractual allocation of risk. I am not surethat the decisions would have been the same if they had.’

69 Above n 2, at 690–91.




begin to protect innocent third parties.70 Therefore, it is difficult to seehow third parties have been prejudiced simply by the abolition of anarguably wider jurisdiction to avoid contracts, albeit one carrying adiscretionary element to avoid prejudicing these self-same third parties.

Perhaps more significantly, it is important to bear in mind that commonmistake operates in a very narrow range of cases, even when freed from thesuffocating grip of Bell v Lever Bros Ltd. It simply has not beendemonstrated that many (or even any) third parties are likely to beadversely affected in circumstances that are demonstrably unfair. In thisrespect, a few considerations must be borne in mind. The clearest examplesof operative common mistakes are res sua and res extincta. But in bothcases, declaring the contract void cannot possibly adversely affect anyinnocent third party purchasers. In the case of res sua, the originalpurchaser already had title and therefore is perfectly capable of passing onthat title to an innocent third party purchaser. In the case of res extincta,the subject matter has ceased to exist so that the question of whether ornot the original purchaser can pass title on to the innocent third partypurchaser becomes moot.

Denning LJ, in inventing the equitable jurisdiction to intervene in Solle vButcher had cunningly drawn upon cases of misrepresentation and unilat-eral mistake.71 In so doing, Denning LJ was able to create the impressionthat the sympathies which tend to lie with innocent third party purchasersin the context of swindlers who impersonate others in order to obtainpossession of goods on credit apply equally in cases of common mistake.However, careful consideration shows that they surely do not. In cases ofmistaken identity, the fraudster typically obtains possession of the goodson credit by impersonating someone else whom the mistaken party iswilling to extend credit to. The fraudster then sells the goods to aninnocent third party purchaser and then either, in the usual case, disap-pears or, more rarely, is caught but found to be insolvent. The same patterndoes not repeat itself in cases of common mistake. The original purchasertypically does not disappear because there is not the same incentive toabscond in cases of common mistake. The innocent third party purchasercan, therefore, generally locate and sue his or her counterparty, the originalpurchaser, should title fail to pass. If the original purchaser should prove tobe insolvent, it is likewise difficult to see why our sympathies should liewith the innocent third party purchaser rather than the seller. In thecontext of mistaken identity, it has been suggested that it is preferable as amatter of legal policy to allocate the loss to the seller who has agreed to sellon credit rather than the purchaser. As Lord Nicholls notes in Shogun

70 A Phang, ‘Common Mistake in English Law: The Proposed Merger of Common Lawand Equity’ (1989) 9 Legal Studies 291.

71 See Solle v Butcher, above n 2, at 690–91.

332 Kelvin F K Low



Finance Ltd v Hudson, ‘it is surely fairer that the party who was actuallyswindled and who had an opportunity to uncover the fraud should bearthe loss rather than a party who entered the picture only after the swindlehad been carried out and who had none’.72

None of these concerns apply in the context of cases of commonmistake. There is no swindler and hence no swindle to discover. Asuperficial reading of Lord Millett in Shogun Finance Ltd v Hudsonappears to suggest a different reason for favouring the innocent third partypurchaser. According to his Lordship, ‘[a]s between two innocent personsthe loss is more appropriately borne by the person who takes the risksinherent in parting with his goods without receiving payment’.73 Takenliterally, and it is far from clear that this was intended by his Lordshipgiven the context in which this passage appears, this view would applywith equal force in the context of common mistake. It is, however, a farless persuasive policy argument than that articulated by Lord Nicholls forsurely the immediate response would be to ask, ‘What of the riskundertaken by the innocent third party purchaser that the original pur-chaser had no title?’ At best, it can be said that both the seller and theinnocent third party purchaser assumed certain risks which have eventu-ated and it is difficult to see why one should be preferred over the other.The truth, therefore, is that the result of nullity is hardly as deleterious tothird party rights as the picture Denning LJ painted, nor is the case forprotection of such rights as may intervene, as obvious as is usuallyassumed.

Finally, it is suggested that if the courts consider that such third partyrights are truly a legitimate concern, there is no logical reason why thecommon law cannot be moulded to address this concern. Cases ofcommon mistake and frustration have for some time been regarded asclosely related. It is today generally acknowledged by the courts thatneither doctrine rests upon the implied intentions of the parties. Suchimplied intention is entirely fictitious.74 In the context of frustration, it hasbeen suggested that concerns over the theoretical basis of the doctrine have

72 [2004] 1 AC 919 (HL) [82].73 Ibid, at [35].74 In the context of frustration: see Davis Contractors Ltd v Fareham Urban DC [1956]

AC 696 (HL) 715 (Viscount Simonds), 720 (Lord Reid) and 728 (Lord Radcliffe). This pointis perhaps best put by Lord Radcliffe who remarked that ‘there is something of a logicaldifficulty in seeing how the parties could even impliedly have provided for something whichex hypothesi they neither expected nor foresaw’ (at 728). In the context of common mistake,the Court of Appeal in The Great Peace (above n 1, at [73]) remarked that ‘the theory of theimplied term is as unrealistic when considering common mistake as when consideringfrustration. Where a fundamental assumption upon which an agreement is founded proves tobe mistaken, it is not realistic to ask whether the parties impliedly agreed that in thosecircumstances the contract would not be binding.’




little practical significance.75 This is not so. So long as the implied termtheory ruled common mistake and frustration, the necessary result of anoperative common mistake would be to render the contract void ab initio(on the basis of an implied condition precedent) whereas the inevitableresult of operative frustration would be to render the contract void as fromthe occurrence of the frustrating event (on the basis of an implied conditionsubsequent). Divorced from the implied term theory, it should be open tothe courts to decide that, instead of common mistake resulting in thecontract being void, it could simply render the contract voidable at theinstance of either party. After all, that is what the courts have suggestedoccurs in equity for common mistake and no one has suggested that itsremedial response, as opposed to the presence of the jurisdiction, ismistaken. The remedy of rescission is certainly not unique to equity sincethe common law exercises precisely this remedy in the context of duress.Why then should it not be available, if deemed desirable, in the context ofcommon mistake?

C. Remedial Flexibility

This leaves then the last concern, which must also be the mostcontroversial—that of remedial flexibility in equity. The obvious firstpoint, and readers must surely be feeling a sense of déjà vu by now, issurely that if remedial flexibility is desirable, why should its availability bedependent on the difficult distinction between a mistake which is ‘funda-mental’ and one which renders the contract ‘essentially different’? Ifremedial flexibility is desirable, then surely the status quo pre-The GreatPeace is not satisfactory, at least not without artful manipulation of thescope of the common law doctrine by the courts. The logical argumentmust again be to allow equity to swallow up the common law so that thereis remedial discretion in all cases of common mistake.76 But before such aview can be acceded to, it must first be determined if the remedialflexibility apparently provided by equity is desirable.

As a matter of authority, its pedigree is certainly suspect. Denning LJpurported to derive his view of remedial flexibility from Cooper vPhibbs,77 where the House of Lords ordered rescission on terms, arguablyon the basis of common mistake.78 However, as the Court of Appeal in TheGreat Peace observed, the House of Lords had not purported to assume abroad discretion to impose terms as Denning LJ seemed to suggest but had

75 GH Treitel, Frustration and Force Majeure, 2nd edn (London, Sweet & Maxwell, 2004)16–013–16–016.

76 Phang, above n 70.77 (1867) LR 2 HL 149.78 Slade, above n 19.

334 Kelvin F K Low



merely imposed terms to protect the rights of parties affected by thedecision.79 The principal objection to a broad discretion to do justicebetween the parties is not that it is not a desirable objective, but that it isan undesirable means to a desirable end. Discretion without direction is aninvitation to uncertainty, arbitrariness and abuse. ‘Doing justice betweenthe parties’ surely cannot be sufficient direction. It is necessary then todistill the particular concerns inherent in the exercise of such discretion.Apart from concerns to protect innocent third parties,80 three otherfunctions of remedial discretion may be discerned from the cases dealingwith common mistake and frustration. First, the remedial discretion wouldallow the courts to reverse any unjust enrichment on the part of eitherparty. Secondly, it may also be used to allocate losses incurred by either orboth parties. Finally, the broad discretion may be used to craft a newagreement for the parties. Each of these functions, when properly exam-ined, will be found to be unnecessary, undesirable or otherwise less thanideal. Furthermore, that the remedial discretion has been sparingly usedwill not come as a surprise given how rarely the courts need to resort to thedoctrine of common mistake. Even when a case of common mistake can beestablished, there is a general reluctance to engage in discretionary revi-sions of rights which would otherwise flow from rescinding the contractsince such consequences are generally just and sound.81

The use of the remedial discretion to reverse any unjust enrichment,whilst arguably useful when Denning LJ first formulated the discretion inSolle v Butcher, is surely today redundant in light of developments in thelaw of unjust enrichment.82

The use of the discretion to allocate losses, whilst not featuring promi-nently in cases of common mistake, has sometimes found expression incases of frustration, in the context of the limited flexibility afforded by theLaw Reform (Frustrated Contracts) Act 1945.83 However appealing it mayinitially appear, the use of discretion to allocate losses is a controversialproposition. The doctrinal basis of such a power is unclear84 and thismakes it difficult to determine when the loss incurred by one party shouldbe shared by the other, and if so, how much should be thus shared. Such a

79 The Great Peace, above n 1, at [105]–[108].80 See text accompanying nn 69–75.81 In West Sussex Properties, above n 32, the Court of Appeal refused the request of the

defendant that rescission in equity be subject to a term that past overpayments in rent shouldbe excluded. The Court of Appeal considered ordinary principles of unjust enrichment indetermining that terms should not be imposed.

82 Which first received formal judicial recognition in the House of Lords in LipkinGorman v Karpnale Ltd [1991] 2 AC 548 (HL).

83 Gamerco SA v ICM/Fair Warning (Agency) Ltd [1995] 1 WLR 1226 (QBD).84 R Stevens, ‘Three Enrichment Issues’ in A Burrows and Lord Rodger (eds), Mapping the

Law: Essays in Memory of Peter Birks (Oxford, Oxford University Press, 2006) 49, 59.




duty to share in the loss would also be, if not anathema to the predomi-nantly adversarial view of contracting adopted by the common law, at leastslightly out of place in a legal system devoid of special pre-contractualduties. If it is felt that such pre-contractual duties ought to be introducedinto the common law, there seems to be no reason why this change needsto be effected in equity rather than at law, now that The Great Peace hasdecisively dissociated the doctrine of common mistake from the impliedterms theory. Perhaps more importantly, any such duty to share is likely toform but a small part of a more significant doctrinal shift in contract lawtheory, not unlike the development of the law of unjust enrichment, andhiding and corralling such a development in the tiny corner of contract lawthat is common mistake would do a disservice to the law.

The most controversial use of the discretion must, however, be its use toredraft the contract for the parties. The first case where this occurred, ofcourse, is Solle v Butcher itself. The facts have been earlier rehearsed andwill not be repeated.85 It will be recalled that the majority of the Court ofAppeal expressed the view that both Solle and Butcher had laboured undera common fundamental mistake which, whilst not affecting their leaseagreement at law, had the effect of rendering the contract voidable inequity. After inventing the remedial discretion in equity, Denning LJ opinedthat the circumstances were such that Butcher would only be permitted torescind the contract in equity upon terms.86 The terms were as follows:Butcher must be prepared to give an undertaking that he would permitSolle to be a licensee of the flat until a new lease was granted. During thisinterim period, Butcher must comply with the procedures which wouldallow him to make additions to the standard rent. Thereafter, if Solle madea written request for a lease, Butcher was obliged to grant him a new leaseon the same terms as the rescinded lease, save in one respect. The rent forthe new lease would be for the full permitted rent, not exceeding £250 perannum. However, Solle was neither obliged to accept the licence nor thenew lease.

The second case in which the equitable discretion was exercised toimpose terms by way of revising the parties’ contract was Grist v Bailey, adecision of Goff J.87 Somewhat simplified, the facts are as follows. Theplaintiff, Frank Grist, had contracted to purchase for £850 a freeholdproperty from the defendant, Minnie Bailey, ‘subject to the existingtenancy thereof’. It transpired, however, that the existing statutory tenantshad passed away so that the market value of the house was £2,250. WhenGrist sued for specific performance, Bailey pleaded rescission for commonmistake. Goff J held that the contract was voidable in equity for common

85 See text accompanying n 15.86 Solle v Butcher, above n 2, at 697.87 [1967] Ch 532.

336 Kelvin F K Low



mistake. As was the case in Solle v Butcher, Goff J was prepared to imposeterms. However, the circumstances leading to Goff J’s readiness to do soare significantly different. It appears that Goff J had not considered theimposition of terms of his own accord but because Bailey had offered tosubmit to a term that rescission should be on condition that she enter intoa fresh contract at a proper vacant possession price. Nor is it clear thatterms were actually imposed because Goff J had merely indicated that hewould impose such a term if required by Grist.88 This hardly amounts toimposition.

Grist v Bailey thus does not offer much support to Denning LJ’s broadequitable discretion to impose terms by way of revising parties’ contracts.This leaves us with only Solle v Butcher and it is a poor champion for thebroad equitable discretion to rewrite terms. At first sight, the case looks apromising one for the imposition of a fresh contract. Solle and Butcherarrived at an agreement of the terms of their contract, and the terms wereineffective only because, by mistake, Butcher had failed to comply with therelevant procedure to have the standard rent increased. Under suchcircumstances, why shouldn’t the courts, in rescinding the contract, simplyimpose the exact same terms upon the parties, after the relevant proceduralsteps have been taken? After all, is it not the case that the contract is beingavoided so as to relieve the parties of the effects of their commonfundamental mistake? If relieving them of their common fundamentalmistake means that they would have entered into a different contract, thenwhy shouldn’t the court impose such a contract on the parties? As anobjective, it is certainly hard to fault. However, its implementation isfraught with difficulties. Cases in which it is clear that the parties wouldhave entered into a particular contract on particular terms if they had beenaware of their shared mistake will be few and far between. Coupled withthe fact that the doctrine of common mistake remains an exceedinglynarrow one, it is questionable if such a case would reveal itself more thanonce in every few decades. Certainly, the curious facts of Solle v Butcherhave not been seen again since 1949 and more than half a century haspassed since. The usefulness of such discretion must be weighed against thetemptation it will pose to judges to extend this discretion beyond such clearcases. Outside of such cases, the discretion to impose a contract upon theparties will be exercised not to relieve the parties of the consequences oftheir mistake but to impose upon them a contract which the courtconsiders fair.

Solle v Butcher itself reveals further difficulties with this use of thediscretion. Although it appears as if the new contract imposed upon Solleand Butcher is in all respects the same as that rescinded by the court, this is

88 Ibid, at 543.




not the case. The rent under this new lease agreement is not £250 perannum but the maximum permitted rent subject to a cap of £250 perannum. Whereas this may be fair to Solle, it is hardly fair to Butcher.Suppose the parties were not mistaken and Butcher had applied foradditions to the standard rent. Suppose additions up to £300 per annumwere permitted, are we to conclude that he would have been content tolease the flat to Solle for £250 per annum? Not only are the terms of thenew contract somewhat lopsided, Denning LJ further weights the condi-tions to rescission in favour of Solle by providing that although Butcher isobliged to grant first the licence and then the new lease to Solle, Solle is notobliged to take up either offer. In short, the terms would be imposed onlyupon Butcher but not Solle should Butcher choose to rescind the contract.The result, of course, is that if there is a catastrophic collapse of the rentalmarket, Solle is free to abandon the rescinded lease agreement and take upa fresh lease elsewhere. Thus, he is protected from any upside that mayfollow from approvals for rent increases beyond the contractually agreed£250 per annum. If it demonstrates anything at all, Solle v Butcher showsthat an unrestrained discretion to impose a contract upon the parties is avery dangerous thing.

Two courses seem open, therefore, so far as imposing new contractsupon parties are concerned. The courts may take the view that they shouldnever be able to impose a contract upon the parties because the advantagesthat such a power would provide in the impossibly rare cases that it will beappropriately exercised are outweighed by the dangers that it poses whenwrongly exercised. If, however, the courts consider that the dangers do notoutweigh the benefits that such a power will bring in those few cases, thenit surely behoves them to articulate the basis upon which the power will beexercised. It is difficult to imagine any other legitimate purpose such apower can serve beyond imposing upon the parties a contract which theywould have made had they not been mistaken at the time of contracting.The difficulty here, of course, is determining when such exceptionalcircumstances exist since it would be akin to locating a needle in ahaystack under the light of the blue moon. Neither option is self-evidentlycorrect though it should be noted that with the dissociation of the commonlaw doctrine from the implied terms theory, there is no reason why such‘discretion’ should not be available at common law. The most appropriateremedy for a common mistake must, in theory at least, surely be to relievethe parties of the effects of that mistake. If it can be shown that the partieswould have entered into a different contract, then it seems perfectlylegitimate for the courts to impose that contract upon the parties as relieffrom common mistake. More often than not, however, this will not bedemonstrable and the more appropriate remedy will be to declare thecontract void.

338 Kelvin F K Low



IV. CONCLUSION

Common mistake after The Great Peace is a very narrow doctrine, but notbecause the courts have crafted a very narrow test. It is a narrow doctrinebecause in most cases where the parties are afflicted by a commonfundamental mistake, the consequences will be addressed by the contractitself, either through its express or its implied terms as to the allocation ofrisk. It is only when the parties’ agreement and the common law’s defaultrules on risk allocation run out that the doctrine of common mistakeoperates. It is demonstrable, and it is hoped has been demonstrated in thisarticle, that many of the concerns over The Great Peace are eitherunfounded or overstated. Furthermore, as a result of focusing on thesewell-rehearsed concerns, the implications for the law of common mistakeas a result of the clear dissociation of the doctrine from the implied termstheory have not been fully appreciated. The Great Peace does not lay downa perfect doctrine of common mistake but it has clarified the law to a verylarge extent and has further laid the foundations for improvements. Ratherthan harping on the past by bemoaning the lost equitable jurisdiction,scholars and judges alike should look to build upon The Great Peace. Ifthey did so, they may find something far more valuable than the flawedequitable jurisdiction that was ‘lost’ out at sea.






14

Contractual Mistake, Intention inFormation and Vitiation: the

Oxymoron of Smith v Hughes

MINDY CHEN-WISHART*

SMITH V HUGHES is a venerable and often cited case which is asfamiliar as it is foundational to students’ understanding of contractlaw.1 Familiarity can impair clarity of vision, but repeated explana-

tions of the case to students can also give rise to a creeping sense of uneasethat something is amiss. A quick recap of the case: a race horse traineragreed to buy oats from a farmer after inspecting a sample which thetrainer mistook for old oats. When the oats turned out to be new and of nouse to him, the buyer refused to pay. The trial judge instructed the jury thatthe buyer could only win on two bases. The jury found for the buyer butdid not state the basis of its decision. On appeal, the court found noevidence for the first basis stated by the trial judge; namely, that the buyerwas not obliged to pay for new oats if the agreement was for old oats. Theseller gave no express or implied warranty of ‘oldness’ and the oatsdelivered corresponded with the sample the buyer had inspected andagreed to buy. That left the trial judge’s second basis; namely, that the sellerknew that the buyer ‘believed, or was under the impression, that he wascontracting for old oats’.2 The trouble was the ambiguity of this directiongiven by the trial judge. It did not differentiate between two types of sellerknowledge, only one of which would void the contract. That is, thecontract is only void if the buyer believed that the seller was promising theoats to be old, but not if the buyer merely believed the oats were old whenthey were not. The lack of evidence for the former (that is the seller’s

* Thanks to John Cartwright, Jane Stapleton, Laura Hoyano, Kate Blackmon andparticipants of the ‘Exploring Contract Law’ symposium at the University of WesternOntario.

1 (1871) LR 6 QB 597.2 Ibid, at 602.



knowledge of the buyer’s mistake as to the promise or as to the terms)suggested to the appellate court that the jury had wrongly voided thecontract for the latter (that is the seller’s knowledge of the buyer’s mistakeas to quality). The verdict being unsafe, a new trial was ordered.

Smith v Hughes is regarded as authority for three propositions:

(i) The law determines contracting parties’ intentions objectively.(ii) However, if one party makes a mistake (that is he or she subjectively

believes something different) about the terms of the contract, thecontract is void if the other party knows of this mistake.

(iii) In contrast, if one party makes a mistake as to fact (that is anunwarranted background assumption) the contract is only void if themistake is fundamental and shared by both parties.

The first proposition seems straightforward and, as we will see, has beentoo easily assumed without sufficient attention to its attributes. But, theinitial plausibility of the second proposition turns into a growing unease oneven fleeting reflection. It requires the student to accept an incoherentpicture of contract law which confidently asserts the dominance of theobjective test in ascertaining contract parties’ intentions, but occasionallypermits a switch to a subjective approach. Indeed, this is the picturepainted by several distinguished commentators3 who state that contractlaw recognises exceptional cases when the subjective approach trumps theobjective approach to void contracts for ‘mistake as to terms’. Forexample, the newest edition of Treitel: The Law of Contract states that theobjective principle ‘is not purely objective’,4 there being ‘three exceptionalsituations, in which the objective principle does not apply, so that the

3 See, eg, GH Treitel, The Law of Contract, 11th edn (London, Sweet & Maxwell, 2003)307 (there are ‘three exceptional situations, in which the objective principle does not apply, sothat the mistake is operative’); J Beatson, Anson’s Law of Contract, 28th edn (New York,Oxford University Press, 2002) 321ff (in some situations ‘the law regards a contract as voidthough at first sight it appears perfectly valid’); AS Burrows, A Casebook on Contract(Oxford, Hart Publishing, 2007) 592 (‘four main exceptions’ to objectivity); PS Atiyah, AnIntroduction to the Law of Contract, 5th edn (Oxford, Clarendon Press, 1995) 84 (in somecases ‘the law abandons the objective interpretation of the first party’s intentions’); EMcKendrick, Contract Law, 7th edn (New York, Palgrave Macmillan, 2007) 23 (‘twosituations in which the objective test is either displaced or modified by a test which at least onthe face of it appears to place greater emphasis upon the subjective intentions of the parties’);E McKendrick, Contract Law: Text, Cases and Materials (Oxford, Oxford University Press,2003) 25 (there are ‘[c]ases in which it has been argued that the courts have resorted to asubjective approach’).

4 E Peel, Treitel: The Law of Contract, 12th edn (London, Sweet & Maxwell, 2007)1–002.

342 Mindy Chen-Wishart



mistake is operative’.5 This characterisation begs the questions: why shouldobjectivity give way in these circumstances and when else might it do so?6

Part I begins to answer these questions by examining the justificationsfor and the meanings of the objective test of contractual intention. I arguethat contract law’s purpose in protecting the autonomy-enhancing institu-tion of contract justifies the objective test. The test determines whether theparties reached agreement and what their agreement was for, by the way itattributes meaning to the parties’ conduct. I argue that the correct versionof the objective test is one that: (a) attributes meaning from the perspectiveof the party observing the conduct; (b) assumes the honesty and reasona-bleness of the party engaged in the conduct; and (c) takes into account thecontext of the parties’ dealings. But, as I further argue, the objectiveapproach is not omniscient. It is not all-seeing and can falter.

Part II then applies this conception of the objective test to the cases ofso-called ‘mistake of terms’ and argues that the latter are merely examplesof this objectivity properly conceived. They can only be described as anexceptional situation of the subjective trumping the objective by assumingan improper test of objectivity. Talk of ‘mistake’ here leads to muddle andconfusion. Moreover, I argue that, on the proper test of objectivity, it islogically impossible to make the fact-finding necessary to void a contractfor ‘known mistake of terms’. To read Smith v Hughes as authority forboth the objective test of intention and for voiding a contract where oneknows of the other’s subjective mistake of term is an oxymoron.

Part III explores the troubling proposition (iii) above. The distinctionbetween a ‘known mistake as to term’ and a ‘known mistaken assumption asto fact’ seems paper-thin, especially when they both relate to the subjectmatter of the contract. In Smith v Hughes, Hennen J said that to void thecontract, ‘the jury should find not merely that the [seller] believed the [buyer]to believe that he was buying old oats, but that he believed the [buyer] tobelieve that he, the [seller], was contracting to sell old oats’.7 Jack Beatsonoffers a clarification of this distinction. According to him, where A sells X apiece of china: the contract stands where ‘X thinks that it is Dresden china. Aknows that X thinks so, and knows that it is not.’8 This is because ‘X’s error isone of motive alone’.9 In contrast, the contract is void where:

X thinks it is Dresden china, and thinks that A intends to contract to sell it asDresden china. A knows that X thinks A is contracting to sell it as Dresden

5 Ibid, at 8–048–8–052. According to Peel, these three exceptional situations are:‘mistake known to the other party’, ‘mistake negligently induced’ and ‘ambiguity’.

6 See H Collins, The Law of Contract, 4th edn (London, LexisNexis, 2003) 235: ‘Theinadequacy of this analysis becomes plain when the textbooks offer lists of exceptions thatdiffer from one another both in their number and character.’

7 Smith v Hughes, above n 1, at 610–11.8 Beatson, above n 3, at 324.9 Ibid.

The Oxymoron of Smith v Hughes 343



china, but does not mean to, and in fact does not, offer more than china ingeneral terms. There is no contract to sell the particular piece of china. X’s error… [was] as to the nature of A’s promise.10

While students might be prepared to accept the theoretical possibility of thedistinction, they struggle with its practical application. Moreover, the justifi-cation for the different remedial thresholds set for different types of mistakesremains unclear. If a mistake as to some non-fundamental quality of thesubject matter (for example that the oats are old) does not negate consent,why should a mistake as to whether the other party promised it to be old beregarded as negating consent? And this is just for starters. How should weunderstand the other categories of mistake (for example mistake of identity,mistake in recording the document, mistake about the nature of the documentsigned and so on) and their own distinctive tests for relief? The student’sdefensive and understandable response is to suspend disbelief, learn off byheart what is not understood and move on quickly in search of firmer ground.

The root of the trouble is the instability of the language used in this areaof the law. The looseness of the key descriptive and prescriptive terms—‘mistake’, ‘defective consent’, ‘objectivity’ and ‘void’—allows quite differ-ent problems to be thrown together resulting in muddle, confusion,obfuscation and loss of precision. To cut through this Gordian knot weneed to identify the correct taxonomy of ‘mistake’ and ‘defective consent’and stabilise the meanings of ‘void’ contracts and the ‘objective’ test ofintentions. My contention is that so-called ‘mistake of terms’ cases areproperly located in stage one of the enforceability question, namely, incontract formation. In contrast, ‘mistake of fact’ cases belong in stage twowhich deals with the vitiation of contract. This reflects the criticaldistinction between disappointed expectations in relation to contractualterms and those in relation to non-terms (‘facts’ or ‘mere representations’),a distinction which is fundamental in the law of misrepresentation andbreach. This classification explains why mistakes as to a contract’s termonly have to be known by the other party while mistakes of fact must beshared and fundamental in order to void contracts. We can also begin totaxonomise the related areas of rectification, non est factum, mistakenidentity and misrepresentation and see the direction in which the coherentresolution of troublesome issues arising under those headings should go.

I. DEFENDING AND DEFINING OBJECTIVITY

Two approaches are generally contrasted in determining the intentions of aparty: the subjective approach refers to a party’s actual intention, regardless

10 Ibid.




of what he or she appears to intend from his or her conduct; and, the objectiveapproach refers to what a reasonable person would interpret as a party’sintention from his or her conduct in all the circumstances. Smith v Hughesaffirmed the dominance of the objective approach. In Lord Blackburn’sfamiliar dicta:

If, whatever a man’s real intention may be, he so conducts himself that areasonable man would believe that he was assenting to the terms proposed bythe other party, and that other party upon that belief enters into the contractwith him, the man thus conducting himself would be equally bound as if he hadintended to agree to the other party’s terms.11

The often stated requirement of consensus ad idem must be filteredthrough this lens of objectivity. The logical corollary is that genuineagreement is unnecessary; the mere appearance of agreement is enough.Any deviation in a party’s actual intention from this objective referencepoint is logically and legally irrelevant.

A. The Justification for Objectivity

The subjective–objective distinction is real. A party’s objective intentionmay, but need not, correspond with his or her subjective intention. Aparty’s conduct may signify consent to a contract when he or she did notreally consent or when he or she believed that they were consenting todifferent terms. Parties can, without any dishonesty, misrepresent their ownmeaning and misinterpret the meaning of others. Why is the objectiveapproach preferred when its effect is to bind parties to contracts that theydid not intend?

Even if contract law is really concerned about the parties’ subjectiveintentions, the parties’ objective intentions may be the best proxy for twopragmatic reasons. The first is one of accessibility. Since ‘the intent of aman cannot be tried, for the Devil himself knows not the intent of aman’,12 the objective approach overcomes the evidential difficulties indetermining what was really in a party’s mind at the relevant time. Thesecond goes to avoidance of fraud. To determine a person’s intentionsimply by reference to what he or she asserts was his or her undisclosedsubjective intention (which may otherwise be impossible to access) is toinvite dishonesty and chaos. Once a conflict has arisen, a person’s incentiveto distort the truth in favour of his own self-interest (even subconsciously)should disqualify such evidence.

11 Smith v Hughes, above n 1, at 607.12 Anon (1478) YB 17 Edw 4, Pasch fo 1, pl 2.




The objective approach, however, is also justifiable on its own terms, forthree further reasons. The first looks at certainty and the protection ofreasonable expectations. Great disruption would ensue if I could escape theliability I appear to have assumed to you by simply asserting that I reallymeant something different. Even if my real intention is objectively provablefrom facts unknown to you, it would be unjust to prioritise it over yourhonest and reasonable, but different, interpretation of my outward con-duct. One of contract law’s main functions is to facilitate the security oftransactions and so enable people to plan and shape their lives on the basisof an apparently enforceable contract. This function would be hopelesslyundermined if legal significance were to attach to the claim that: ‘When Isay “white” I mean “black.”’ The objective approach allows parties toknow in advance how their own conduct will be interpreted and how theyare entitled to interpret the conduct of others. This overlaps with and isreinforced by the next justification.

The next justification is based on the protection of the autonomy-enhancing institution of contracting. The purpose of contract law is toprotect the practice of undertaking voluntary obligations because it enablesparties to act autonomously, to make their own arrangements, to shapetheir own lives.13 The objective test of intentions is one of the rules ofengagement necessary to protect the integrity of the contracting processand to prevent its abuse.14 Holding parties to the objective standard: (a)prevents them from reneging on their undertakings; (b) gives them strongincentives to take care not to misrepresent their own intentions (eveninnocently) nor to misinterpret the intentions of others; and (c) extends thepractice beyond ongoing relationships where it would otherwise not exist:‘But for the support of the law, contracts between complete strangerswould not be as numerous and common as they are.’15

Finally, objectivity is intrinsic to contracting. Making a contract isessentially an exercise in the communication of choice, and communicationis impossible without objectivity.16 We have to suspend our own meaning,enter imaginatively into the other’s world and ask: ‘What meaning do theythink they’re conveying?’ and ‘What will they think I am meaning?’17 AsMay LJ said in Ove Arup v Mirant Asia Pacific Construction Ltd:‘Subjective intention or understanding, unaccompanied by some overt

13 J Raz, ‘Promises in Morality and Law’ (1982) 95 Harvard Law Review 916, 933.14 Ibid, at 928–38.15 Ibid, at 934.16 T Endicott, ‘Objectivity, Subjectivity, and Incomplete Agreements’ in J Horder (ed),

Oxford Essays in Jurisprudence (4th Series) (Oxford, Oxford University Press, 2000) 151; DGoddard ‘The Myth of Subjectivity’ (1987) 7 Legal Studies 263.

17 A Kronman, ‘Paternalism and the Law of Contracts’ (1983) 92 Yale Law Journal 763.




objectively ascertainable expression of that intention or understanding, isnot relevant.’18 Similarly, Ernest Weinrib explains that objectivity is intrin-sic to private law since:

Interaction between free wills engages the external aspect of practical reason …On stepping into a world of interaction, the freely willing actor establishes apresence there through acts that have an externally recognizable nature. Purelymental imaginings and reservations, however real they are to the actor orhowever serious the consequences to which they might in due course lead, haveno status in this world of interaction. Thus criminal wrongdoing requires anactus reus; contract cannot be held hostage to the vagaries of a private intention;and the claim to property must involve some act in the world of appearances …The external nature of action implies a world of shared social meanings. Onlywithin such a world can juridical acts by each of the parties be interpreted froma perspective common to both and thus have significance as external acts.19

Intention is wholly dependent on manifestations interpreted in a context ofshared meaning. The existence and extent of contractual obligations aredetermined by the signs made—the moves in the language game beingplayed. Any legal concern with undisclosed intention is senseless: itcontradicts the very idea of contract as an agreement between parties whoconvey and receive meaning.

B. Objectivity and Voluntariness

The objective test only bites when it yields a result that trumps a party’sactual and different intention. This fact is highlighted by those advancinghybrid principles of contractual liability based on reliance, unjust enrich-ment, fairness, and public policy20 over one based on the parties’ voluntaryexercise of autonomy. Objectivity undoubtedly undermines the view ofcontract law based on the importance of promise-keeping.21 But, it isconsistent with the view that one of contract law’s primary functions is toprotect the facilitative institution of contracting.22 That is, ‘to protect both

18 [2004] BLR 49 (CA) [62].19 E Weinrib, The Idea of Private Law (Cambridge, Harvard University Press, 1995) 104.20 See, eg, G Gilmore, The Death of Contract (Columbus, Ohio State University Press,

1974); P Atiyah, The Rise and Fall of Freedom of Contract (Oxford, Clarendon Press, 1979);P Atiyah, Essays on Contract (Oxford, Clarendon Press, 1986).

21 C Fried, Contract As Promise (Cambridge, Harvard University Press, 1981) 61: ‘the courtimagines that it is respecting the will of the parties by asking what somebody else, say the ordinaryperson, would have intended by such words of agreement … it palpably involves imposing anexternal standard on the parties … clearest in cases of what is called unilateral mistake.’

22 Raz, above n 13, at 937: ‘To enforce voluntary obligations is to enforce morality throughthe legal imposition of duties on individuals. In this respect it does not differ from the legalproscription of pornography.’ For a similar view, see Endicott, above n 16; Goddard, above n 16;H Sheinman, ‘Contractual Liability and Voluntary Undertaking’ (2000) 20 OJLS 205.




the practice of undertaking voluntary obligations and the individuals whorely on that practice… One protects the practice of undertaking voluntaryobligations by preventing its erosion—by making good any harm causedby its use or abuse.’23 The objective principle prevents people from abusingthe practice of contracting by making it appear that they have agreed toobligations when they have not. It is consistent with the harm principle24

which holds that the only proper purpose for imposing legal obligations onindividuals is to prevent harm. But unlike reliance theorists who focus onharm to the individual, Joseph Raz extends the notion to include ‘institu-tional harm’. Liability is not imposed to protect voluntariness on theindividual level (except at the very high threshold level of non estfactum),25 but in order to protect the practice of undertaking voluntaryobligations; to preserve its appeal and utility.

Thus, voluntariness, albeit on an institutional level, remains the distinc-tive touchstone of contractual liability. The objective test of contractformation is not an embarrassment to the view that the purpose of contractlaw is to support the autonomy-enhancing practice of undertaking volun-tary obligations. Paradoxically, it is in order to protect the practice ofcontracting from debasement that the law recognises the validity ofcontracts that are not voluntary obligations.26

C. Defining Objectivity

A contract is comprised of the coincidence of the parties’ voluntaryintentions to be bound by certain terms. How do we know what each partyintended? How is the meaning of another person constructed, communi-cated, interpreted, or accessed?27 Securing the priority of objectivity oversubjectivity is only the starting point. Three further questions arise: (i)Objectivity from whose perspective? (ii) Objectivity on what standard? (iii)Objectivity on what evidence?

(i) Objectivity from Whose Perspective?

William Howarth sets out three perspectives from which a party’s intentionhas been assessed by the courts.28 The first is detached objectivity,

23 Raz, ibid, at 933.24 JS Mill, ‘On Liberty’ in Three Essays: On Liberty, Representative Government, The

Subjection Of Women (London, Oxford University Press, 1975).25 See text accompanying nn 106–15.26 Sheinman, above n 22; Raz, above n 13, at 935.27 See further C Dalton, ‘An Essay in the Deconstruction of Contract Doctrine’ (1985) 94

Yale Law Journal 997, 1039–65; J Finnis, ‘The Priority of Persons’ in J Horder (ed), OxfordEssays in Jurisprudence (4th series) (Oxford, Oxford University Press, 2000) 1, 11–13.

28 W Howarth, ‘The Meaning of Objectivity in Contract’ (1984) 100 LQR 265.




sometimes called ‘fly on the wall’ objectivity. This perspective adopts theviewpoint of a reasonable person independent of that of either contractparty. The second perspective is ‘promisor’ objectivity, which adopts theviewpoint of the reasonable person seeking to avoid the contract (that isthe defendant). The third perspective is that of ‘promisee’ objectivity,which adopts the viewpoint of the reasonable person seeking to enforce thecontract (that is the claimant). The terminology of ‘promisor’ and ‘prom-isee’ objectivity is problematic because they depend entirely on the rela-tively insignificant fact of which party initiates proceedings; indeed, inrespect of the term in contention, either the actual promisor or promiseecould be the ‘promisor’ (that is the defendant) or ‘promisee’ (that is theclaimant) on Howarth’s scheme. Moreover, even if we use ‘promisor’ and‘promisee’ in their natural sense, each party is both in a bilateral contract:‘[h]e is a promisor with regard to what he undertakes to perform and apromisee with regard to what he is entitled to receive’.29

In light of these difficulties, the more transparent distinction is between‘actor objectivity’ (what a reasonable person in the actor’s position wouldmean by her conduct), and ‘observer objectivity’ (what a reasonable personin the observer’s position would interpret the actor’s conduct as meaning).Both interpret the actor’s conduct, but they may differ because the factualmatrix informing each party’s perspective may differ; facts affecting ‘actorobjectivity’ may not be apparent to the observer and vice versa. Forexample, the observer (seller) may reasonably interpret the actor’s (buyer’s)conduct as meaning ‘oats’ when the actor has provable reasons, unknownto the observer, for reasonably intending to mean ‘old oats’. Whichperspective—the detached person’s, the actor’s or the observer’s—should beadopted?

Detached objectivity is implied by Lord Blackburn’s appeal to what ‘areasonable man would believe’.30 It is strongly supported by various dictaof Lord Denning31 and some academic writing.32 This approach couldpromote certainty and protect third parties who rely on the apparentmeaning of contractual documents. But it does not address the main

29 J Vorster, ‘A Comment on the Meaning of Objectivity in Contract’ (1987) 103 LQR274, 276–78.

30 Smith v Hughes, above n 1, at 607.31 Solle v Butcher [1950] 1 KB 671 (CA) 691: ‘[O]nce the parties, whatever their inmost

states of mind, have to all outward appearances agreed with sufficient certainty in the sameterms on the same subject matter … [n]either party can rely on his own mistake to say it wasa nullity from the beginning, no matter that it was a mistake which to his mind wasfundamental, and no matter that the other party knew that he was under a mistake.’ See alsoFrederick E Rose (London) Ltd v William H Pim Junior and Co Ltd [1953] 2 QB 450 (CA)461 [Rose v Pim].

32 See, eg, M Furmston, Cheshire Fifoot & Furmston’s Law of Contract, 15th edn(Oxford, Oxford University Press, 2007) 306: ‘the question is not what the parties had intheir minds, but what reasonable third parties would infer from their words or conduct’.




justifications for objectivity (protecting parties’ reasonable expectationsand reliance, and the practice of undertaking voluntary obligations).33 Itmay even facilitate, rather than prevent, fraud; a party could insist on thedictionary meaning of the words used even when he or she knows that thatwas not the agreement. The exceptions to the parole evidence rule tellagainst it, most notably the doctrine of non est factum.34 As John Spencerobserves, while ‘[i]t may be acceptable for the law occasionally to forceupon one of the parties an agreement he did not want . . . surely there issomething wrong with a theory which forces upon both of the parties anagreement which neither of them wants’.35 Actor objectivity does notencounter this objection and moreover, is consistent with the first twojustifications for objectivity (accessibility and avoidance of fraud); thequestion would be one of the availability of evidence. However, it fallsshort in protecting the observer’s reliance and expectations and, conse-quently, the institution of contracting; it is also inconsistent with the natureof communication as the conveying of meaning to the observer.

It is observer objectivity which has the weight of authority36 and ticksall the justifications for objectivity. It also receives reinforcement from anunlikely source. The doctrine of consideration justifies the observer, whohas ‘paid’ for the promised performance, in adopting a reasonable viewof the actor–promisor’s conduct from his or her (the observer’s) perspec-tive. The actor who accepts ‘payment’ is bound by the observer’sreasonable interpretation of his or her conduct. Actors cannot rely ontheir different meaning, even if they can prove its reasonableness fromtheir perspective.37

33 Atiyah, Rise and Fall, above n 20, at 663; JR Spencer, ‘Signature, Consent, and theRule in L’Estrange v Graucob’ [1973] CLJ 104, 110, 112.

34 See text accompanying nn 106–15.35 Spencer, above n 33, at 113.36 See André & Cie SA v Marine Transocean Ltd (The Splendid Sun) [1981] 1 QB 694

(CA) as explained in Paal Wilson & Co v Partenreederei Hannah Blumenthal (The HannahBlumenthal) [1983] 1 AC 854 (HL) 865 [The Hannah Blumenthal]; Chaloner v Bower (1984)269 EG 725 (CA); Tankreederei Ahrenkeil GmbH v Frahuil SA (The Multibank Holsatia)[1988] 2 Lloyd’s Rep 486 (QBD) 493; OT Africa Line Ltd v Vickers Plc [1996] 1 Lloyd’s Rep700 (QBD Comm) [OT Africa Line]; Freeman v Cooke (1848) 2 Exch 654, 154 ER 652, 653;Vorster, above n 29, at 278, 283.

37 Centrovincial Estates plc v Merchant Investors Assurance Co Ltd [1983] Com LR 158(CA) 24 [Centrovincial Estates]; OT Africa Line, ibid, at 703. This is the case unless theparties are genuinely at cross-purposes as to the subject matter of the contract and the latter isso ambiguously expressed that the court cannot determine which version is more likely, as inRaffles v Wichelhaus (1864) 2 Hurl & C 906, 159 ER 375 (discussed below, in textaccompanying nn 101–5). And see, R Stevens, ‘Objectivity, Mistake and the Parole EvidenceRule’ in A Burrows and E Peel (eds), Contract Terms (Oxford, Oxford University Press, 2007)101, 102–103.




The contrast is with bare promises contained in a deed38 where actor-objectivity has priority. Millett J explains in Gibbon v Mitchell that thedeed will be set aside for mistake ‘if the court is satisfied that the disponerdid not intend the transaction to have the effect which it did’.39

(ii) Objectivity on What Standard?

Consistent with the aim of facilitating the practice of contracting andpreventing its abuse, the objective approach is infused with a bias in favourof the just and reasonable interpretation, and against the unjust andunreasonable.40 It treats a contract party as an honest and reasonableperson who will not take an unjust view of the other party’s intentions, norgive a dishonest or misleading view of his or her own intentions. The sameidea underlies Hugh Collins’ reference to the need for ‘clean hands’ and hisidentification of a ‘duty to negotiate with care’ and to ‘refrain fromunconscionable conduct’.41 This view is consistent with John Finnis’‘conversational’ model of interpretation, which prioritises the observer’sinterpretation, when it is subject to the qualifier that:

[A] properly juridical interpretation will not be as ready to consider authoritativean unjust as it will a just meaning. Thus it differs from sensible conversational-ists, who like good historians are quick to detect, and not too ready to overlook,interlocutors’ perhaps vicious purposes and deficiencies of personal character.42

(iii) Objectivity on What Evidence?

How much evidence should be taken into account in determining how ajust and reasonable observer would have interpreted the actor’s conduct?Two versions of objectivity can be detected lying at opposite ends of aspectrum: the traditional formal objectivity and the more recent contextualobjectivity. Formal objectivity is closely related to detached objectivity; itseverely limits the evidence to be taken into account and prioritises themaccording to a fairly strict hierarchy of probative value. Thus, signed finalwriting contained in a contractual document is the best evidence ofintention (hence the signature rule, the parole evidence rule and the generaleffectiveness of entire agreement clauses). It is superior to unsigned final

38 Lady Hood of Avalon v Mackinnon [1909] 1 Ch 476. The claimant settled £8,600 onboth her daughters having forgotten that she had previously settled an even larger sum on theelder daughter. Her gift was set aside.

39 [1990] 1 WLR 1304 (Ch D) 1309.40 Raz, above n 13, at 934–5: ‘because the predominant purpose of contract law is to

support existing moral practices, one might expect that the conditions of the validity ofcontracts will reflect common moral conceptions … to mirror common moral views’.

41 Collins, above n 6, at 234–7.42 Finnis, above n 27, at 13.




writing contained in a contractual document (the ‘ticket cases’) or con-tracts formed by exchange of writing. This in turn is superior to speech,which is superior to non-verbal conduct (a nod, a wink, contractualperformance), which, finally, is superior to silence or omissions.43

Formal objectivity represents a relatively depersonalised interpretation ofconduct as opposed to a genuine search for the meaning of the person engagedin the conduct.44 The temptation to formal and detached objectivity isstrongest where the contract is evidenced in some document as was the case inL’Estrange v F Graucob Ltd45 and Butler Machine Tool Co v Ex-Cell-OCorp.46 These cases are criticised for their unrealistic analysis of the factsbecause the courts prioritised the signatures over other contradictory evi-dence of the signer’s intention, which was apparent to the observer.47 LordDevlin observed that the signature rule, particularly in the context of standardform contracts, is premised on a ‘world of make-believe which the law hascreated’.48 Signatures are still treated as binding the signatory almost abso-lutely, as if ‘some kind of magic operated to take the contract out of the usualrules that govern the formation of contracts’.49

Formal objectivity continues to hold sway, particularly with commercialcontracts where certainty is rightly prized and a signature can reasonablybe taken to manifest intention to consent to its contents. But a moreexpansive contextual objectivity is growing in influence, especially innon-commercial contexts. The Canadian decision of Tilden Rent-a-Car Cov Clendenning applied it to a hurried consumer transaction with finelyprinted, unexpected and harsh clauses.50 The Ontario Court of Appealheld that ‘the signature by itself does not truly represent an acquiescence to

43 Silence generally does not constitute acceptance even if it was clearly intended: seeFelthouse v Bindley (1862) 6 LT 157 (CP); but inaction or silence may amount to intention toabandon one’s claims if accompanied by reliance: see The Hannah Blumenthal, above n 36, at865; Allied Marine Transport Ltd v Vale Do Rio Doce Navegacao SA (The Leonidas D)[1985] 1 WLR 925 (CA).

44 See Dalton, above n 27, at 1039–65; A De Moor, ‘Intention in the Law of Contract,Elusive or Illusory?’ (1990) 106 LQR 632, 635–55; Goddard, above n 16.

45 [1934] 2 KB 394 (Div Ct). The court upheld a signed document excluding all the seller’sliability although the seller knew that the buyer did not read the ‘regrettably small print’.

46 [1979] 1 WLR 401 (CA). The facts were as follows. B offered to sell a machine on itsterms including a price variation clause, but E placed an order on different terms excludingprice variation. The latter prevailed because B had signed and returned the tear-offacknowledgement slip on E’s order form stating ‘we [B] accept your order on the terms andconditions stated thereon’ (ibid, at 403). The court ignored B’s covering letter sent with thetear-off slip insisting on their original terms (ie including the price variation clause).

47 See, eg, Spencer, above n 33, at 121–2.48 McCutcheon v David MacBrayne Ltd [1964] 1 WLR 125 (HL) 133.49 Spencer, above n 33, at 117.50 (1978) 83 DLR (3d) 400 (Ont CA). Thus, the hirer of a car was not bound by a clause

making him liable for damage to the car in a wide variety of circumstances going far beyond whathe would reasonably expect since: (a) he had paid extra for additional insurance coverage; (b) itwas plain that he did not read nor was expected to read the fine print; and (c) the speed with whichthe transaction was completed was one of the attractions of the service being offered.




unusual and onerous terms which are inconsistent with the true object ofthe contract’ unless the business had taken reasonable steps to draw theconsumer’s attention to these terms.51 The court cited Stephen Waddams:

One who signs a written document cannot complain if the other party reason-ably relies on the signature as a manifestation of assent to the contents, orascribes to the words he used their reasonable meaning. But the other side of thesame coin is that only a reasonable expectation will be protected. If the partyseeking to enforce the document knew or had reason to know of the other’smistake the document should not be enforced.52

In English law, the general shift away from literalist interpretations ofcontractual documents towards a more contextual interpretation53 culmi-nates in Lord Hoffmann’s judgment in Investors Compensation SchemeLtd v West Bromwich Building Society.54 Accordingly, the court shouldplace itself in the same factual matrix as the parties and to take intoaccount absolutely everything55 reasonably available to them which wouldhave affected the way that they interpret the contractual documents and,logically, any other manifestations of intent.56 This approach largelydiscards the ‘old intellectual baggage of “legal” interpretation’ replacing itwith ‘the common sense principles by which any serious utterances wouldbe interpreted in ordinary life’.57 The search is for the meaning of theperson, and not merely of the conduct. Consistently with this view, Finnisreasons that since law is ‘for the sake of persons and its rules arefundamentally relationships between persons’ the intent of persons shouldbe treated as if it really matters.58 Raz opposes a philosophical individual-ism which fails to recognise that an act has different normative implica-tions depending on its social context.59 Similarly, Kronman argues against

51 Ibid, at 407.52 Ibid, at 405, citing SM Waddams, The Law of Contracts (Toronto, Canada Law Book,

1977) 191.53 See, eg, Prenn v Simmonds [1971] 1 WLR 1381 (HL) 1383–4 (Lord Wilberforce): ‘In

order for the agreement … to be understood, it must be placed in its context. The time haslong passed when agreements … were isolated from the matrix of facts … and interpretedpurely on internal linguistic considerations. We must inquire beyond the language and seewhat the circumstances were with reference to which the words were used. … English law[was not] left behind in some island of literal interpretation.’

54 [1998] 1 WLR 896 (HL) 912–13 [West Bromwich]. See also Lord Napier and Ettrickv RF Kershaw [1999] 1 WLR 756 (HL) 763 (Lord Steyn): ‘Loyalty to the text of a commercialcontract, instrument or document read in its contextual setting is the paramount principle ofinterpretation … [T]he court ought generally to favour a commercially sensible construction[because the] commercial construction is likely to give effect to the intention of the parties.’

55 See criticisms of this approach raised by C Staughton, ‘How do the Courts InterpretCommercial Contracts?’ [1999] CLJ 303.

56 See Lord Nicholls, ‘My Kingdom for a Horse: the Meaning of Words’ (2005) 121 LQR577 criticising the rules excluding evidence of previous negotiations and subsequent conduct.

57 West Bromwich, above n 54, at 912.58 Finnis, above n 27, at 11–13.59 Raz, above n 13, at 931–2.




a clinical and impoverished approach which treats persons as denatured,disembodied egos, in favour of a context-infused richness of meaning.60

The contextual approach is exemplified by Mannai Investment Co Ltd vEagle Star Life Assurance Co where the tenant wrote ‘12th’ when his‘evident intention’ was ‘13th’.61 Lord Hoffmann explained that words donot in themselves refer to anything; it is people who use words to refer tothings and it ‘is a matter of constant experience that people can conveytheir meaning unambiguously although they have used the wrong words’.62

His Lordship added:

It is of course true that the law is not concerned with the speaker’s subjectiveintentions. But the notion that the law’s concern is therefore with the ‘meaningof his words’ conceals an important ambiguity. The ambiguity lies in a failure todistinguish between the meanings of words and the question of what would beunderstood as the meaning of a person who uses words . . . This involvesexamining not only the words and the grammar but the background as well. So,for example, in Doe d Cox v Roe … the landlord of a public house in Limehousegave notice to quit ‘the premises which you hold of me . . . commonly called . . .The Waterman’s Arms’… [T]here were no such premises in the parish ofLimehouse. But the tenant did hold premises of the landlord called TheBricklayer’s Arms . . . The meaning was objectively clear to a reasonablerecipient, even though the landlord had used the wrong name . . . There was noneed to resort to subjective meaning.63

In the decision of the Court of Appeals of New York in Utica CityNational Bank v Gunn, Cardozo J agreed that the meaning of words ‘isnot always the meaning of the parties’.64 In that case, Gunn (G) and othersgave a guarantee of $115,000 ‘on the consideration of’ Utica City NationalBanks’ (UCNB) ‘loans and discounts.’ Cardozo acknowledged that: ‘Thislooks to the future. It excludes the past,’ so that the words construed alonewould support G’s refusal to pay absent a new loan from UCNB. Yet,evidence of the genesis and aim of the guarantee showed that both partieswere well aware that it was to secure an existing loan of that preciseamount to the principal debtor. Cardozo J therefore interpreted ‘loans anddiscounts’ to include ‘renewals’ consistent with the ‘efficacy and purpose’of the transaction. He preferred to ‘give a new shade of meaning to a wordthan to give no meaning to a whole transaction’.

60 Kronman, above n 17.61 [1997] AC 749 (HL). The tenant had written ‘January 12th’ in the notice to terminate

the lease when the lease stipulated that he only had power to do this on its anniversary whichwas January 13th. The House of Lords held that the notice was effective to determine thelease on January 13th.

62 Ibid, at 774.63 Ibid, at 775.64 222 NY 204 (1918) 208.




The equitable doctrine of rectification is further support for the priorityof observer-contextual objectivity over detached-formal objectivity. Theparole evidence rule barring appeal to evidence of intention outside thecontractual document does not apply where rectification is sought since theclaim is precisely that the contractual document does not reflect theparties’ agreement as the coincidence of each party’s evident intention.65

Moreover, where the parties agree on the meaning of a particular phraseused in the contractual document, the contract can be rectified to makeclear that the phrase bears the meaning agreed.66 Rectification is alsoallowed where (a) one party knows of the other party’s mistake about thecontents of the contractual document and of the mistaken party’s realintentions,67 (b) fails to draw the mistaken party’s attention to the mistake;and, (c) the mistake benefits the unmistaken party or prejudices themistaken party.68 Logically, the case for rectification should be no weakerif one party’s misrepresentation has induced that shared belief in themeaning of words used in the document, even if innocent.69 In Curtis vChemical Cleaning & Dyeing Co Ltd, Curtis (C) took a wedding dress toChemical Cleaning & Dyeing (CCD) for cleaning and was asked to sign a‘receipt’ exempting CCD from liability for ‘any damage howsoever aris-ing’.70 C did so after being assured that CCD’s exemption was only inrelation to any damage to beads or sequins on the dress. The dress wasreturned badly stained. The court concluded that CCD was not allowed torely on the wide exclusion clause; it could only rely on the exemption tothe limited extent represented.

The crucial distinction is not between subjectivity and objectivity, sincesubjectivity is simply irrelevant; but rather between detached-formal objec-tivity on the one hand, and observer-contextual objectivity on the other.The former might necessitate resort to the notion of exceptional ‘subjectiv-ity’ to explain the outcome of cases where contracts are voided for‘mistakes of terms’, but this is unnecessary on the more expansiveobserver-contextual objectivity.71

65 Lovell and Christmas Ltd v Wall (1911) 104 LT 85 (CA) [Lovell and Christmas]. Eventhe presence of an ‘entire agreement’ clause in the contract does not prevent rectification: seeJJ Huber (Investment) Ltd v Private DIY Co Ltd [1995] NPC 102 (Ch D) [JJ Huber].

66 London Weekend Television v Paris and Griffith (1969) 113 SJ 222 (HC). See also ReButlin’s Settlement Trusts [1976] Ch 251.

67 Actual knowledge is required, but this includes wilfully shutting one’s eyes to theobvious, or wilfully and recklessly failing to make such inquiries as an honest and reasonableperson would have made. The doctrine also applies to one who has made ‘false andmisleading statements’ to divert the other from discovering the mistake, see Commission forthe New Towns v Cooper [1995] 1 Ch 259 (CA) 280 [New Towns v Cooper].

68 Thomas Bates & Sons Ltd v Wyndhams Ltd [1981] 1 WLR 505 (CA) 515–16 [Bates vWyndhams], 520–21; New Towns v Cooper, ibid.

69 See to the contrary Rose v Pim, above n 31.70 [1951] 1 KB 805 (CA).71 See, eg, the examples in n 3 above.




II. APPLICATIONS OF OBJECTIVITY IN CONTRACT FORMATION

The cases often cited as exceptional resorts to subjectivity (allowingcontracts to be voided for mistake as to terms) rest on the erroneousreference point of detached-formal objectivity. In fact, they are straightapplications of observer-contextual objectivity; talk of ‘knowing of theother party’s mistake’ is unnecessary and logically impossible.

A. Was there an Objective Agreement?

According to Smith v Hughes the components of an operative mistake ofterms are:

(i) There is an objectively determinable agreement (A1).(ii) The actor mistakenly believes the agreement is not for A1 but rather

for a different agreement (A2).(iii) The actor also mistakenly believes the observer to be agreeing A2.(iv) The observer knows of the actor’s mistakes in (ii) and (iii).

My argument is that once the objective agreement A1 is found viaobserver-contextual objectivity, the observer (in Smith v Hughes, the seller)cannot logically know of the actor’s (in Smith v Hughes, the buyer)intention to agree to A2. The observer cannot honestly and reasonablybelieve that the actor intends to agree to A1 (implicit in finding anagreement for A1) and honestly and reasonably believe that he or sheintends to agree to the inconsistent A2. While, in real life, it may beaccepted that we can simultaneously hold contradictory beliefs, that canhardly be, and is not, the approach of the law which must necessarily comedown on one side or the other. It is worth taking time over the proof.

(i) The Objective Point of Reference

Was the contract for ‘oats’ as the seller alleged, or ‘old oats,’ as the buyeralleged? This question breaks down into two: (a) What did each party’sconduct, in the factual matrix of the case, honestly and reasonably lead theother to believe about his or her intention? (b) Did the parties’ objective(ostensible) intentions coincide? The court’s approach and conclusion wasentirely consistent with observer-contextual objectivity. The agreement wasfor ‘oats’ period; the seller gave no express or implied warranty of ‘oldness’and the buyer offered a price after inspecting a sample of oats whichcorresponded with what was later delivered.72 Conversely, the seller had no

72 Smith v Hughes, above n 1, at 603, 605 (Cockburn CJ), 607 (Blackburn J), 609(Hennen J).




reason to know of the buyer’s intention to buy old oats; the buyer claimedthat trainers only used old oats as a rule, but the buyer had since sold newoats to a trainer;73 and while the price was high for new oats, oats werethen very scarce and expensive.

(ii) Buyer’s Mistaken Assumption about the Subject Matter of theContract

The buyer’s unilateral mistaken assumption that the oats he was buyingwere old is legally irrelevant in itself.74 Even the seller’s knowledge of thebuyer’s mistake does not change this, absent a misrepresentation or ageneral obligation of disclosure. Passive acquiescence in another’s self-deception does not void the contract.75 The market system rewardsresearch and knowledge so that a knowledgeable party must generally beallowed to take advantage of a less knowledgeable (mistaken) party.Cockburn CJ gave two examples of this phenomenon: a sale of land whichthe buyer knows that the seller is ignorant of the existence of a mine underit,76 and the sale of a horse ‘as is’, where the seller knows that the buyermistakenly believes the horse to be sound. Both contracts are binding.77

(iii) The Buyer’s Mistake as to Terms

Although the seller gave the buyer no reason to believe that he waspromising old oats, the buyer may have assumed this to be true on someindependent basis, such as from a third party’s misrepresentation.78 Whilethe buyer ‘believing he has purchased old oats’ as opposed to merely‘believing the oats he purchased were old’ will, in practice, be difficult todistinguish, it is a distinction routinely made in the context of liability formisrepresentation and breach.79

(iv) The Seller Knows of the Buyer’s Mistake of Terms

Here is the nub. Even if we find that the buyer mistook the seller’s offer, itwill be impossible to find that the seller knew of it. No rational partywould concede such knowledge when knowledge of the other’s mistake as

73 Ibid, at 602.74 See the discussion of operative mistaken assumption of fact at text accompanying nn

146–50.75 Smith v Hughes, above n 1, at 603.76 Ibid, at 604.77 Ibid, at 606. See also, ibid, at 607 (Blackburn J).78 There may be an action against the third party for fraud or negligent misrepresenta-

tion.79 See text accompanying nn 136–8.




to subject matter can be admitted without cost (as in (ii)). Absent such anadmission, the question is whether that knowledge can be inferred fromthe seller’s own conduct or on an honest and reasonable interpretation ofthe buyer’s conduct. The crucial point is that this is just another way ofasking the same question as (i), namely, ‘what is the objective contract?’The same body of evidence is interrogated, but a contradictory answer to(i) is ostensibly required to void the contract. In fact, four answers arepossible, none of which necessitates or indeed is susceptible of thedescription ‘known mistake of term’.

First, if the buyer’s conduct is honestly and reasonably interpreted by theseller as consenting to buy ‘oats’ and vice versa80 then that is theobjectively determined contract at (i). The seller cannot, at the same time,honestly and reasonably believe the buyer to be consenting to (that theseller was promising) ‘old oats’. The seller cannot simultaneously believethe buyer to be consenting to two contradictory things.

Second, if the buyer’s conduct is honestly and reasonably interpreted bythe seller as promising ‘old oats’, and vice versa, then that is the objectivelydetermined contract at (i). The objective reference point has simply moved.As Hannen J said:

If … the [seller] knew that the [buyer], in dealing with him for oats, did so on theassumption that the [seller] was contracting to sell him old oats, he was awarethat the [buyer] apprehended the contract in a different sense to that in which hemeant it, and he is thereby deprived of the right to insist that the [buyer] shall bebound by that which was only the apparent, and not the real bargain.81

On this scenario, the ‘real bargain’ must be for old oats. Not only can thebuyer refuse payment for the new oats, he can, in principle, enforce thecontract for old oats. Although this is a theoretical possibility, there was nobasis for such a conclusion on the facts of Smith v Hughes.82

Third, if the seller’s conduct merely induced the buyer’s mistake of factthat the oats were old (that is it was not a term promising old oats), thecontract is voidable for misrepresentation.83 Fourth and finally, if the seller

80 The seller’s conduct is honestly and reasonably interpreted by the buyer as consentingto buy ‘oats’.

81 Smith v Hughes, above n 1, at 610 (emphasis added).82 Ibid, at 611 (Hennen J): ‘It may be assumed that the [buyer] believed the oats were old,

and it may be suspected that the [seller] thought he so believed, but the only evidence fromwhich it can be inferred that the [seller] believed that the [buyer] thought that the [seller] wasmaking it a term of the contract that the oats were old is that the [buyer] was a trainer, andthat trainers, as a rule, use old oats; and that the price given was high for new oats, and morethan a prudent man would have given.’ However, the seller was ignorant of trainers’ buyinghabits and had subsequently sold new oats to a trainer; oats were also scarce at the time andtherefore commanded higher prices than had traditionally been the case.

83 Ibid, at 605 (Cockburn CJ): ‘If, indeed, the buyer instead of acting on his own opinion,had asked the question whether the oats were old or new, or had said anything whichintimated his understanding that the seller was selling the oats as old oats, the case would




neither promised nor represented the oats were ‘old’, but still had reason toknow that the buyer did not intend to contract on his terms (simply ‘oats’)but without knowing what the buyer was intending to contract for, therewould simply be no agreement between the parties.84

Where parties are at cross purposes on the terms of the contract, theobjective test of intention, properly understood will yield an answer to allcases, supplemented by the doctrine of misrepresentation. There is either acontract on one party’s meaning or the other’s, with any such contractpossibly voidable for misrepresentation, or there may simply be noobjectively corresponding offer and acceptance. Non est factum aside,there is no work left for a separate mistake of term doctrine to do. WhileSmith v Hughes undoubtedly recognises the category of ‘known mistakesas to terms’, the case itself is not a positive instance of category. The actualdecision was only to order a new trial to see whether the jury’s conclusionto void the contract could be defended in the light of a more precisestatement of the qualifying conditions. The judges clearly did not thinkso.85 On retrial, the buyer could only win if further evidence emerged toshow that there was actually a contract for ‘old oats’, an actionablemisrepresentation as to the age of the oats, or no objective agreement atall.

Two examples can be given where there is questionable or no objectivecoincidence of intentions necessary for contract formation. In Centrovin-cial Estates plc v Merchant Investors Assurance Co Ltd, land had beenrented at £68,320 per annum.86 When the rent was reviewed, the landlordproposed £65,000 when he meant to say £126,000; the tenant immediatelyaccepted. The tenant was given leave to defend the landlord’s action for adeclaration that no binding contract had been concluded. The Court ofAppeal held that the figure of £65,000 would stand unless the landlordcould prove that the tenant knew or ought reasonably to have known thatits offer was unintended when they purported to accept it. In Chwee Kin

have been wholly different; or even if he had said anything which shewed that he was notacting on his own inspection and judgment, but assumed as the foundation of the contractthat the oats were old, the silence of the seller, as a means of misleading him, might haveamounted to a fraudulent concealment, such as would have entitled the buyer to avoid thecontract.’

84 Ibid, at 610 (Hannen J): ‘If by any means he knows that there was no real agreementbetween him and the promiser, he is not entitled to insist that the promise shall be fulfilled ina sense to which the mind of the promiser did not assent.’ See also, ibid, at 607 (BlackburnCJ).

85 Blackburn J (ibid, at 608) could ‘not see much evidence to justify a finding for thedefendant … if the word ‘old’ was not used. There may have been more evidence than isstated in the case; and the demeanour of the witnesses may have strengthened the impressionproduced by the evidence there was; but it does not seem a very satisfactory verdict if itproceeded on [the mistake] ground.’ Hannen J (ibid, at 611) also found ‘very little, if any,evidence to support a finding upon [the mistake ground] in favour of the [buyer].’

86 Above n 37.




Keong v Digilandmall.com Pte Ltd, Digilandmall.com’s (D’s) employeemistakenly advertised a commercial laser printer for $66 on D’s website(less than 2 per cent of the actual retail price of $3,854).87 By the time theerror was detected, 4,086 orders had been received and confirmation notesautomatically dispatched within a few minutes. D resisted Chwee’s (C’s)action to enforce their order for 1,606 printers, alleging that its unilateralmistake was known to C. The Singapore Court of Appeal upheld thefinding that the buyers were ‘fully conscious that an unfortunate andegregious mistake had indeed been made by the defendant’.88 Althoughactual knowledge of the mistake was said to be required, a very generousview was taken of its scope. According to the court, it can be inferred fromcircumstantial evidence: ‘Phrases such as “must have known” or “couldnot reasonably have supposed” are really evidential factors or reasoningprocesses used by the court in finding that the non-mistaken party did, infact, know of the error made by the mistaken party.’89 Moreover, itincludes ‘“Nelsonian knowledge,” namely, wilful blindness or shuttingone’s eyes to the obvious’.90 In both these cases, the issue was whether theparties reached agreement and what the agreement was. Talk of mistake ofterms known to the other party merely imports unnecessary distraction.

The equitable analogue is Webster v Cecil.91 In that case, Cecil (C)offered to sell land for £1,250 when he intended to say £2,250. Themistake must have been obvious to Webster (W) because C had previouslyrefused to sell for £2,000 and C informed W of his mistake immediatelyafter W’s purported acceptance. The court of equity refused W’s claim forspecific performance, but, in the absence of sufficient evidence to hold thebuyer to a contract at £2,250, common law would have found noconcluded contract at all.

B. What was the Objective Agreement For?

In contrast to cases where objective agreement is absent, where D knows ofC’s objective meaning and leads C to believe that D is consenting to it, there isa contract on C’s meaning although this deviates from the literal or ordinary

87 [2005] SGCA 2, affirming [2004] 2 SLR 594 (HC) [Digilandmall.com].88 According to the High Court, the relevant factors were: (i) the ‘stark gaping difference

between the price posting and the market price of the printers’; (ii) the fact that the buyerswere ‘well-educated professionals—articulate, entrepreneurial and, quite bluntly, streetwiseand savvy individuals’; and (iii) the fact that the printers were purchased in the ‘dead of night’with ‘indecent haste’ and the e-mails between the purchasers showed that they were anxiousto place orders before the mistake was corrected: see Digilandmall.com (HC), ibid, at[142]–[145].

89 Digilandmall.com (CA), above n 87, at [35].90 Ibid, at [42].91 (1861) 30 Beav 62, 54 ER 812.




meaning of C’s conduct. In Hartog v Colin & Shields the seller offered to sell3,000 Argentine hare skins at a 10d ‘per pound’ when he really meant and hadpreviously offered 10d ‘per piece’.92 There were three pieces to the pound.The buyer purported to accept and sued for damages when the seller refusedto deliver for one third of his intended price. The court’s conclusion in favourof the seller has been explained in terms of the seller’s subjective intentiontrumping his objective intention when the buyer knows of the seller’s mistakeas to terms (that is knows of the seller’s subjective intention): he is preventedfrom snatching a bargain known not to have been intended for him. Again,this rests on the erroneous detached view of objectivity.93 On the correctobserver-contextual version of objectivity, an honest and reasonable buyerwould interpret the seller as meaning ‘per piece’. The court found that thebuyer ‘could not reasonably have supposed’ the seller intended to quote theprice ‘per pound’ given the practice in the trade and the pre-contractualnegotiations (verbal and written), which always discussed the price ‘per piece’and never ‘per pound’.94 Moreover, Singleton J said that he found it ‘difficultto believe that anyone could receive an offer for such a large quantity ofArgentine hares at a price so low as 3d per piece without having the gravestdoubts of it … the plaintiff must have realised, and did in fact know, that amistake had occurred’.95 If the observer knows or has reason to know that theactor’s meaning does not coincide with the detached objective interpretationof his or her conduct (that is has made a mistake in stating the terms of his orher offer) then, as a reasonable and just person, the observer would not treatthe other party as having agreed to it. An unreasonable and unjust observer isnot allowed to assert their unreasonableness and ‘snap up’ the offer. Indeed, inHartog v Shields, it is clear that the buyer knew that the seller meant to offer‘per piece’, an offer which he purported to accept. Against that point ofreference, the seller made no mistake. In principle, he could have, but chosenot to, enforce the contract on a ‘per piece’ basis.96 The decision was simplythat he was able to resist the buyer’s enforcement of a contract on the ‘perpound’ basis since there was no such contract.

92 [1939] 3 All ER 566 (KBD) [Hartog v Shields].93 See, eg, McKendrick, Contract Law: Text, Cases and Materials, above n 3, at 40:

‘objectively, the parties to reach [an] agreement. The defendants offered to sell their hareskinsat a price per pound and the plaintiff accepted that offer. It was the fact that the plaintiffknew that the defendants were mistaken that led Singleton J to conclude that no contract hadbeen concluded to sell their hareskins at a price per pound. This suggests that the vital factorin persuading Singleton J to conclude that the plaintiff was not entitled to recover damageswas his finding that the parties were not subjectively agreed.’

94 Hartog v Shields, above n 92, at 568.95 Ibid, at 567–8.96 This is consistent with the requirements of rectification for unilateral mistake as to the

wording of a document. This analysis is found in text accompanying nn 68–71. See Bates vWyndhams, above n 68, at 515–16, 520–21; New Towns v Cooper, above n 67.




Scriven v Hindley can be interpreted in the same way.97 It also instancesthe objective approach’s bias to the just and reasonable. The buyersuccessfully bid, but refused to pay, for an auction lot he believed tocontain hemp when it actually contained the much cheaper commoditytow. Lawrence J found no contract on the unnecessarily wide basis that‘the parties were never ad idem as to the subject-matter of the proposedsale’, one intending to sell tow and the other intending to buy hemp.However, here, the non-coincidence of the parties’ ‘subjective’ meanings,which was unknown to either party, would have been irrelevant98 withoutthe seller’s misleading auction catalogue. This described the goods as somany bales in different lots, all bearing the same shipping marks which,witnesses explained, never happened before for different commodities fromthe same ship. The bidder, quite reasonably, did not to foresee the potentialfor confusion: being only interested in buying hemp, he had only inspectedthe hemp on show and not the tow bearing the same shipping marks.Lawrence J held that since the confusion was deliberately perpetrated bythe seller-observer to swindle the bank financing the shipment, ‘it waspeculiarly the duty of the auctioneer to make it clear to the bidder . . .which lots were hemp and which lots were tow’.99 While auctioneers aregenerally entitled to assume that bidders know what they are bidding for(their mistakes are legally irrelevant), they cannot do so if they havecarelessly, albeit unintentionally, induced the bidder’s mistake. The case isauthority for observer-contextual objectivity, not for the priority of subjec-tive intention over detached-formal objectivity. It shows that where theseller’s conduct gives the bidder reason to believe that he or she is biddingfor hemp the seller cannot treat the bidder’s offer as for tow. Indeed, thereis arguably a contract for hemp.

The equitable analogue is Denny v Hancock.100 In that case, thepurchaser inspected the property with the assistance of plans that showedone side to be bounded by trees. He naturally concluded that threemagnificent trees going up to an iron fence were inside the property; in factthey were not, the real boundary being denoted by stumps made incon-spicuous by shrubs. In denying specific performance, James LJ stressed thatany reasonable prospective purchaser would be misled. He said: ‘[i]f I haddone exactly what this gentleman did, and taken their plan in my hand,and gone through the property … I should have arrived at exactly the sameconclusion as this gentleman did.’ Although the focus was on whether thepurchaser could resist specific performance, it is arguable that if there wasany objective agreement, it was for the property as the seller knew that the

97 [1913] 3 KB 564, 568.98 Robinson, Fisher and Harding v Behar [1927] 1 KB 513.99 Scriven v Hindley, above n 97, at 569.

100 (1870) LR 6 Ch App 1, 11.




buyer would have understood and intended to contract for. At commonlaw, the buyer should have been entitled to compensation for his loss ofexpectation.

C. Latent Ambiguity

In the well-known case of Raffles v Wichelhaus the parties contracted tobuy and sell goods ‘to arrive ex Peerless from Bombay’, the buyer intendingthe ship Peerless arriving in October while the seller delivered on anothership, also called Peerless, arriving in December.101 The court upheld thebuyer’s refusal to pay but without giving its reasons. The case is interpretedas an exception to the objective test, either on the basis that the parties’subjective intentions do not coincide102 or that there is a mutual mistake inthat each party is mistaken as to the other’s intention.103 With respect,both explanations are unnecessarily wide.

Exact subjective coincidence of intentions must be rare in contracting;non-correspondence is not generally a ground for voiding the contract. Theobjective test, by definition, makes irrelevant any party’s subjective mistakeabout the other party’s intention. What matters is each party’s evidentintention honestly, reasonably and contextually interpreted from theobserver’s perspective. On this approach, offer and acceptance corre-sponded; there was agreement to buy and sell goods ‘to arrive ex Peerlessfrom Bombay’. The problem was identifying which of the two ships fittingthat description was the subject of the agreement, which version should beenforced. The answer was ‘don’t know, can’t say’. It belongs with othercases where vagueness (rather than lack of agreement) prevented contractformation, as where agreements to sell goods ‘on hire-purchase terms’104

or ‘subject to war clause’105 but the courts could not say which of themany different, and each reasonable, versions of such terms the partiesintended. The parties were objectively agreed, but the agreement sufferedfrom latent ambiguity which was impossible to resolve by reference to thecontext: objectivity simply ‘ran out’. Even the more expansive contextualobjectivity is not omniscient.

101 Raffles v Wichelhaus, above n 37.102 See Beatson, above n 3, at 321–2 who discusses the case under the heading of ‘Absence

of genuine agreement’ and ‘Offer and acceptance not coincident’.103 See Furmston, above n 32, at 306.104 Scammell and Nephew Ltd v Ouston [1941] AC 251 (HL).105 Bishop & Baxter v Anglo-Eastern Trading and Industrial Co Ltd [1944] KB 12 (CA).




D. Actor Objectivity: Non est factum

The non est factum doctrine comes closest to appearing to take account ofcontract parties’ subjective intentions. Where one party’s mistake as to thenature of a signed contractual document is ‘fundamental’, or ‘essential’, or‘radical’, or ‘very substantial’, or ‘serious,’106 non est factum can void thecontract even if the mistake is unknown to the other party (if it was, therewould be no objective agreement as in Smith v Hughes) and has not beeninduced by the other party’s misrepresentation (if it was, the contractwould be voidable even if the mistake was not ‘fundamental’).107 But nonest factum is not an instance of the exceptional priority of subjectivity; thedoctrine represents an exceptional switch from the perspective of theobserver to that of the actor, although the actor in question must suffer therelevant cognitive disability108 and not have been careless.109 The test isstill objective since the claimant must show that his mistaken belief aboutthe nature of the document was honest and reasonable for someone withthat disability in the circumstances.

The doctrine is justified on the basis of no consent,110 but this isover-inclusive. Specifically, what justifies the switch from the observer tothe actor’s perspective? The answer is that while contract law aims tofacilitate the autonomy-enhancing institution of contract by preventing itsabuse (hence observer-objectivity which safeguards voluntariness at theinstitutional level),111 contract law cares about autonomy at the individuallevel and hence actor-objectivity, to this extent. The actor’s conduct which

106 Saunders v Anglia Building Society [1971] 1 AC 1004 (HL) [Saunders], affirmingGallie v Lee [1969] 2 WLR 901 (CA).

107 See United Dominions Trust Ltd v Western [1976] QB 513 (CA) [United DominionsTrust]. In United Dominions Trust, the defendant’s hire purchase agreement, tainted by thedealer’s fraud, would have been void against the finance company although the latter wasignorant of the fraud, had the defendant not been careless in signing a blank agreement andhad the agreement been fundamentally different from that intended by the defendant.

108 Saunders, above n 106, at 1016: be ‘permanently or temporarily unable through nofault of their own to have without explanation any real understanding of the purport of aparticular document, whether that be from defective education, illness or innate incapacity,’or from being tricked.

109 In Saunders, ibid, a widow who had broken her glasses was barred by her carelessnessin not checking the identity of the transferee. The requirement of care would disqualify aclaimant who signs a document in blank leaving another to fill in the details, see UnitedDominions Trust, above n 107.

110 See Foster v Mackinnon (1869) LR 4 CP 704. At 711, Byles J said the contract isinvalid ‘on the ground that the mind of the signer did not accompany the signature: in otherwords, that he never intended to sign, and therefore in contemplation of law never did sign,the contract to which his name is appended’. See also Hasham v Zenab [1960] AC 316 (PC)335; Mercantile Credit Co Ltd v Hamblin [1965] 2 QB 242 (CA) 268, 280. Saunders, aboven 106, at 1026.

111 See text accompanying nn 14–16.




counts as consenting to a contract must be: (a) intentional;112 (b) accom-panied by knowledge that the conduct will count as contracting;113 and (c)accompanied by a minimum threshold of accuracy as to the gist of thecontract.

In relation to the first criteria, an action will count as intentional if theactor knowingly engages in the conduct. Thus, in this area the law imposesa very weak form of intent which covers almost everything short ofautomatism, sleep walking, being pushed and so on.

In relation to the second criteria, knowledge that the conduct will countas contracting is absent if, for example: (a) the actor believes he or she isgiving someone an autograph when the actor is actually signing a contract;or (b) the actor drives into a car park believing it to be free when it ischarged, even if the other party reasonably believes the actor is consentingto a contract.

In relation to the third criteria, effective consent does not requireaccuracy on every detail of the content (hence, the observer-objective testof intention, the signature rule, the rules on incorporation and the curingof uncertainty in formation and so on). Nevertheless, if a claimant canshow that he or she has made a fundamental mistake as to the very natureof the obligation undertaken, then the claimant can claim that the consentwas not meaningful enough to bind him or her. The same reasoningjustifies voiding a contract for common fundamental mistake as to fact(assumption) since this radically defeats the purpose or the means forachieving the purpose of the contract.114 Of course, a very high thresholdof seriousness is required (the difficulty of defining and applying this isnotorious). Other hurdles may be imposed to protect the certainty andsecurity required of a valuable institution of contract. But the idea is that acontract law which enforces voluntarily assumed obligations should, atleast, accept that a party who is so fundamentally mistaken about the gist,or core, or substance of what he or she has undertaken (he or she is noteven in the right ‘ball park’) has not really undertaken anything. Phraseddifferently, you need not actually agree to every rule of the club to bebound by them, but you must at least have joined the right club.115

112 Goddard, above n 16; Sheinman, above n 22.113 Endicott, above n 16; Sheinman, above n 22.114 See text accompanying nn 156–8.115 See further De Moor, above n 44, at 635–55; Endicott, above n 16, at 151; Raz, above

n 13, at 933; Sheinman, above n 22.




E. Mistaken Identity

What of the ‘notoriously unsatisfactory’116 English law on mistake as tothe other party’s identity? The central problem is the rogue posing assomeone else who buys goods with a worthless cheque or on credit, sellsthe goods on and then absconds before his or her fraud is discovered.Should the original owner be able to recover the goods from an innocentthird-party purchaser? The plethora of potentially applicable rules, theusual presence of misrepresentation and signatures and the concern toprotect innocent third-party purchasers have yielded a set of potentiallycontradictory and ossifying guiding rules. Thus, the contract is said to be:(a) void if the mistake is as to the other party’s identity but voidable if themistake is only as to that party’s’ attribute; and (b) void if the contract ismade ‘in writing’,117 but voidable if it is made ‘face-to-face’.118 These rulesof thumb wrongly suggest that the contract is void if the claimant is verymistaken (the claimant wins), but voidable if he or she is only a littlemistaken (the third party wins).

The proper starting point must be the basic rule of contract formationthat no one can accept an offer which he or she knows or has reason toknow is not intended for that party. Neither can a party rely on anapparent acceptance knowing it was in response to an offer believed tohave come from someone else.119 If the other party makes an offer oraccepts an offer under mistake as to the offeror’s identity, the question isone of formation—whether a contract has been made at all—not whetherthe mistake should vitiate any prima facie valid contract. This assumes thatthe identity of the other contract party is important to the claimant.120 If itis not, (for example, retailers are not generally concerned about theidentity of the shopper, or the auctioneer with the identity of the bidder),121

the contract comes into existence despite the mistaken identity although itmay be voidable if it was nevertheless induced by the rogue’s misrepresen-tation.

116 Shogun Finance Ltd v Hudson [2004] 1 AC 919 (HL) [1], [34] (Lord Nicholls). For asimilar view, see Law Reform Committee, 12th Report: Transfer of Title to Chattels (London,Her Majesty’s Stationary Office, 1967); G McMeel, ‘Interpretation and Mistake in ContractLaw: “The Fox Knows Many Things…”’ [2006] Lloyd’s Maritime and Commercial LawQuarterly 49; C MacMillan, ‘Rogues, Swindlers and Cheats: The Development of Mistake ofIdentity in English Contract Law’ [2005] CLJ 711; C MacMillan, ‘Mistake as to IdentityClarified?’ (2004) 120 LQR 369; C Hare, ‘Identity Mistakes: A Missed Opportunity?’ (2004)67 MLR 999; AL Diamond, ‘Law Reform Committee: Twelfth Report on the Transfer of Titleto Chattels’ (1966) 29 MLR 413.

117 Cundy v Lindsay (1878) 3 App Cas 459 (HL).118 Phillips v Brooks [1919] 2 KB 243.119 Boulton v Jones (1857) 2 H & N 564, 157 ER 232 (Ex Ct).120 Ingram v Little [1961] 1 QB 31 (CA) 57; Lewis v Averay [1972] 1 QB 198 (CA) 209.121 Dennant v Skinner [1948] 2 KB 164.




However, even if vital to the claimant, the identity of the other party tothe contract does not quite fit the description of ‘term’; it is artificial to saythat the rogue promises his identity as a term of the contract. Nevertheless,while the identity of one’s contract partner does not go to what thecontract is for, it does go to the equally important question of who onechooses to contract with (to whom one has chosen to undertake contrac-tual obligations). In this sense it is very much part of the contractformation question. We can regard it as a condition precedent, a term ofthe claimant’s offer, or of his or her acceptance. On this view, andconsistently with observer-contextual objectivity, claimants should be ableto deny contracts if they can show that: (i) they were mistaken about theother party’s identity; (ii) that identity was vital to them;122 and (iii) theother party (the rogue) knew or had reason to know of (i) and (ii).

Proof of (ii) would be analogous to that required to show that astatement made during negotiations is a term or collateral term of thecontract, rather than a mere representation.123 Identity is clearly vitalwhere, for example, an offer is made only to persons fitting particulardescriptions which exclude the rogue (‘current students of a particularuniversity’, or being ‘over 18 years of age’); or, where the rogue knowsfrom previous dealing that the claimant is unwilling to contract with himor her (since he or she is barred from a pub or a soccer match).124 Theclaimant must show that ‘but for’ the mistake the claimant would not haveentered the contract. In contrast, a misrepresentation is operative if it wasmerely a reason for the claimant entering the contract. On Lord Millett’sscenario in Shogun Finance Ltd v Hudson, the contract is voidable where aman books a hotel room ‘for himself and a girlfriend under a common butfictitious name in order to give the impression (when such things mattered)that they were married’.125 But the contract may be void if the man knowsthat the hotel would only accept married couples. Logically, mistake aboutany attribute of the rogue which is vital to the claimant should count.However, his creditworthiness should be excluded because if the claimantallows payment by cheque or extends credit, this is simply a business riskthat it takes.

As for (iii), where the rogue has misrepresented his or her own identity,this is weighty evidence that the rogue not only knows of the claimant’smistake (which he or she has induced), but also that the rogue knows ofthe importance of his or her identity to the claimant (hence the incentive to

122 Ingram v Little, above n 120, at 57; Lewis v Averay, above n 120, at 209.123 Heilbut Symons & Co v Buckleton [1913] AC 30 (HL) [Heilbut Symons]; Oscar Chess

Ltd v Williams [1957] 1 WLR 370 (CA); City of Westminster Properties (1934) Ltd v Mudd[1959] Ch 129 [Westminster Properties v Mudd], and see text accompanying nn 145–7.

124 Said v Butt [1920] 3 KB 497.125 Above n 116, at [78].




lie). Boulton v Jones shows that the rogue may have the relevant know-ledge even without an active misrepresentation.126 The determinationshould be made without undue distortion by the presumptions thatcontractual intention is present when a rogue’s misrepresentation occurs inthe claimant’s presence (face-to-face), but absent when it is made in theclaimant’s physical absence (and the contract is constructed from anexchange of writing or is reduced to a written document). If a party wasignorant that the claimant had mistaken his or her identity or of theimportance of this identity to the claimant then the contract comes intoexistence (for example, I enter a contract with you only because Imistakenly believe you are a famous actor). A party may be similarlyignorant even if he or she has misrepresented this identity. For example, if,in the hotel scenario, the man can show that his lie is not primarily aimedat deceiving the hotel but to mislead, say, journalists or his wife. Thatleaves voidness due to common (shared) fundamental mistaken assump-tion, but this requires a factual scenario which is impossible to imagine(since the rogue knows his or her own identity) without some medicalcondition akin to amnesia.

On this approach, mistaken identity cases should be resolved viacontract formation analysis. If this approach is taken, many more caseswould result in a conclusion that there was no contract. The concern withthird party rights, a concern which rightly belongs in property law, shouldbe dealt with separately and not be allowed to twist contract law principleswhich are aimed at assessing the rights between the contracting parties.The tail should not wag the dog.

III. DISTINGUISHING MISTAKEN ASSUMPTIONS

A. Imposing Taxonomic Order: Formation and Vitiation

Smith v Hughes leaves us with a final conundrum: why is it that the buyer’smistaken belief that the oats he purchased were old attracts no relief(because it is not shared by the seller nor sufficiently fundamental inimportance), but the buyer’s mistake that the seller had promised the oatsto be old would void the contract if known to the seller? If a mistake as tosome non-fundamental quality of the subject matter does not negateconsent, why should a mistake as to whether the other party promised itdo so? Indeed, one might go on to ask how we should understand the

126 Above n 119. In that case, J sent a written order for some goods to Brocklehurst, whoJ had dealt with previously and against whom he could set off sums that Brocklehurst owedhim. B took over Brocklehurst’s business and filled J’s order without disclosing the change ofownership.




different requirements of the formidable list of different mistakes com-monly identified in case law and commentary.127

The problem is one of unstable classification. The categories of mistakecut across each other; some are about who has made the mistake, someabout the type of mistakes made, some about the seriousness of themistake, and some about the different requirements for relief under thecommon law and at equity (for example, knowledge, fault on the defend-ant’s part or lack of fault on the claimant’s part). The same scenario maybe susceptible to different descriptions to which different legal conse-quences attach. Absence of a dominant taxonomy in the law of contractualmistake impedes understanding, leaving the student or judge to take refugein a list approach.

Clarity about the underlying structure of a subject is vital to clearthinking, principled development of the law and the eradication ofinconsistencies. As Lord Steyn reminds us, ‘in law classification is impor-tant. Asking the right questions in the right order reduces the risk of wrongdecisions.’128 A model of mind-boggling taxonomy which conforms to nosingle classifying scheme is the categorisation of animals identified by JorgeLuis Borges in ‘a certain Chinese encyclopaedia’. There:

animals are divided into (a) belonging to the emperor, (b) embalmed, (c) tame,(d) suckling pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in thepresent classification, (i), frenzied, (j) innumerable, (k) drawn with a very finecamelhair brush, (l) et cetera, (m) having just broken the water pitcher, (n) thatfrom a long way off look like flies.129

What classificatory rationale underpins different types of mistakes and soexplains the different requirements for relief? To unravel the tangle weneed to start with another capacious idea in play here; namely, that whencontracts are set aside, including for mistake, it is because of the claimant’s‘defective consent’ to the contract. Like mistake, ‘defective consent’ is notprecise enough to be useful, but is loose enough to cause real trouble. HLAHart warned that the process of reasoning which holds that since consentgets you into contract, only lack of consent will get you out is ‘a disastrous

127 A non-exhaustive list of mistakes includes: common mistake, unilateral mistake,mutual mistake, cross purpose mistake, mistakes going to the root of the contract, absence ofgenuine agreement, common law mistake, equitable mistake, mistake as to the identity orintention of the other party, fundamental mistake as to the identity, ownership, existence,quality or quantity of the subject matter, fundamental mistaken assumption, mistakenrecording of the contract (rectification), and fundamental mistake about the nature of thedocuments signed (non est factum).

128 Attorney-General v Blake [2001] 1 AC 268 (HL) 290.129 M Foucault, The Order of Things (New York, Pantheon Books, 1970) xv, referred to

in D Johnston and R Zimmermann, ‘Unjustified Enrichment: Surveying the Landscape’ in DJohnston and R Zimmermann (eds), Unjustified Enrichment: Key Issues in ComparativePerspective (Cambridge, Cambridge University Press, 2002) 3, 25–6.




over-simplification and distortion’ of the law governing the vitiation oftransactions.130 This reasoning fails to recognise that consent is a necessarybut not sufficient, condition of contractual liability, and that the enforce-ability of a contract is a two-stage inquiry. Even when the language ofconsent is used in both the formation and vitiation stages, they deal withqualitatively different concerns. ‘Formation’ asks whether the parties havesucceeded in reaching agreement. Here, the commitment and contentquestions are merged in the offer and acceptance approach. In contrast,‘vitiation’ asks whether, in view of the relevant circumstances, a claimantshould be released from liability under the contract, despite her consent toit.131

As Hart explains:

When the student has learnt that in English law there are positive conditionsrequired for the existence of a valid contract … his understanding of the legalconcept of a contract is still incomplete … For these conditions, althoughnecessary, are not always sufficient and he has still to learn what can defeat aclaim that there is a valid contract, even though all these conditions are satisfied.That is the student still has to learn what can follow on the word ‘unless’ whichshould accompany the statement of these conditions … the law has a wordwhich with some hesitation I borrow and extend: This is the word ‘defeasible’used of a legal interest in property which is subject to termination or ‘defeat’ ina number of different contingencies but remains intact if no such contingenciesmature. In this sense then, contract is a defeasible concept.132

The truth of these statements is obscured by those who take the languageof impaired consent as suggesting ‘that there are certain psychologicalelements required by the law as necessary conditions of contract and thatthe defences [vitiating factors] are merely admitted as negative evidence ofthese’, rather than as ‘a compendious reference to the defences with whichclaims in contract may be weakened or met’.133 In short, contract forma-tion is not conditional on the parties’ minds being free from mistake; theparty seeking to enforce a contract does not have the onus of proving theabsence of mistake on the other’s part. The claimant’s success in voiding acontract for mistake is not proof of lack of consent to contract formation.

130 HLA Hart, ‘The Ascription of Responsibility and Rights’ (1948) 49 Proceedings of theAristotelian Society 171, 183 (reprinted in AGN Flew (ed), Logic and Language—First Series(Oxford, Blackwell Publishing, 1963)).

131 Ibid, at 174.132 Ibid, at 174–5 (emphasis in the original).133 Ibid, at 177. At 180, Hart explains that ‘the logical character of words like

“voluntary” are anomalous and ill-understood. They are treated in such definitions as wordshaving positive force, yet, as can be seen from Aristotle’s discussion in Book III of theNicomachean Ethics, the word “voluntary” in fact serves to exclude a heterogeneous range ofcases such as physical compulsion, coercion by threats, accidents, misstakes, etc, and not todesignate a mental element or state; nor does “involuntary” signify the absence of this mentalelement or state.’




In these cases, the contract is set aside despite consent because the presenceof, for example, duress134 or undue influence135 means that the law doesnot ascribe the normal responsibility it would to the victim’s consent. Hisor her consent is only deemed to be defective. Talk of defective consent inthe context of vitiating factors is conclusory, not explanatory; it is, as Hartreminds us, merely a shorthand for the variety of factors rendering atransaction defeasible.

B. Terms and Assumptions

The primary taxonomic distinction between (a) contract formation and (b)vitiation of contract is fundamental. It mirrors that key distinction betweenterms inside a contract and assumptions (non-terms, or ‘mere representa-tions’) outside the contract. The distinction can be exceedingly difficult todraw.136 As Smith v Hughes shows, the quality of the subject matter (thatthe oats are old) may be either a term or an assumption, attracting verydifferent tests for relief. The same applies to the existence,137 or identity138

of the subject matter of the contract.The distinction between terms and facts is vital in determining whether

false statements in pre-contractual negotiations attract the consequences ofbreach of contract or for misrepresentations inducing agreement to thecontract. Since misrepresentations are but induced mistakes, it stands toreason that the distinction should be equally important in the law ofmistake.

(i) Mistake of Terms and Mistakes of Fact

Mistake of terms and mistakes of fact raise entirely different issues andinvoke different principles. So-called ‘mistake as to terms’ goes to theformation and contents of the contract. These cases raise the issues of offer

134 This resonates with the rejection of the ‘overborne will’ explanation of duress inUniverse Tankships Inc of Monrovia v International Transport Workers Federation [1983] 1AC 366 (HL). As Lord Scarman said at 400: ‘The classic case of duress is … not the lack ofwill to submit but the victim’s intentional submission arising from the realisation that there isno other practical choice open to him.’

135 See Royal Bank of Scotland v Etridge (No 2) [2002] 2 AC 773 (HL) [7] where LordNicholls said that whenever the defendant’s procurement of the claimant’s consent is judgedimproper by the standards of the undue influence doctrine, that consent will not be deemedan expression of the claimant’s will.

136 See Heilbut Symons, above n 123; Oscar Chess Ltd v Williams, above n 123.137 It was classed as a term in McRae v Commonwealth Disposals Commission [1951] 84

CLR 377 (HCA) and has been interpreted as a fact in Couturier v Hastie (1856) 5 HL Cas673.

138 It was classed as a term in Scriven v Hindley, above n 97, at 567 and as a fact in Leafv International Galleries [1950] 2 KB 86 (CA).




and acceptance (objectively determined), implied terms and collateralterms. Moreover, breaches of contractual terms trigger forward-lookingremedies in the nature of vindicating expectations.

Mistakes of fact are mistaken assumptions about the context in whichthe contract is made. These mistakes affect a party’s motivation for, orassessment of, the desirability of the contract. ‘Mistakes of fact’ go to thevitiation of contract. It asks whether—despite the parties’ objective consentto the contract—one party should be excused from contractual liability.The remedy is backward looking and, broadly speaking, aimed at restoringthe parties to their pre-contractual position.

(ii) Mistake and Objectivity

The word ‘mistake’ connotes deviation from the accurate (unmistaken)point of reference. Talk of ‘mistake’ makes sense when applied to assump-tions, less so to contractual terms. A party makes a mistaken assumption ifhis or her evident (and in that sense ‘objective’) belief about the matterdeviates from the independently verifiable truth of the matter. You believethe oats you physically inspected and bought are old but they are not. Ibelieve that the sea vessel I hired to standby and evacuate the crew of mydistressed ship pending the arrival of a rescue tug is 35 miles away; in factit is 400 miles away.139 You lease a room to watch the Royal Coronationprocession, but it has already been cancelled.140 I pay to terminate yourcontract of employment not realising that I could have done so withoutpaying.141

In contrast, so-called ‘mistake of terms’ are not so straightforward.Unlike facts, the reference point of contractual terms has no independentlyverifiable existence. It is the construct of the parties and emerges from thecoincidence of each party’s objective interpretation of the other’s intention.In Smith v Hughes the evidence pointed to an agreement for ‘oats’. Sincethe seller had reason to believe that that was the buyer’s intention, he couldnot simultaneously know (that is have the inconsistent belief) that thebuyer really intended to buy ‘old oats’. The buyer’s belief is treated as oneof factual assumption, which only voids the contract if it is fundamentaland shared by the seller.

Even where a contractual document purporting to contain the parties’agreement is present, the question is whether each party has given the otherreason to believe that he or she intends to be bound by the four corners ofthe document, hence the exceptions to the signature rule and the paroleevidence rule and the remedy of rectification. If not, the court may be able

139 Great Peace Shipping v Tsavliris [2003] QB 679 (CA) [The Great Peace].140 Griffith v Brymer (1903) 19 TLR 434 (KBD).141 Bell v Lever Brothers Ltd [1932] AC 161 (HL).




to add implied or collateral terms that may even override the terms of thecontractual document;142 or may rectify the contractual document to bringit into line with the parties’ evident and agreed intention.143 In each case,the parties’ real agreement, if provable, trumps the different written recordof their agreement. Thus, the contractual document will only be theobjective reference point for determining mistake if, objectively deter-mined, the parties have agreed that it embodies their agreement and theiragreement has not been wrongly recorded. The resulting potential forcircularity should not obscure the fact that in ascertaining whether therehas been a mistake of terms and who has made it, it is not necessarilyenough to just look for deviation from any contractual document. Thepoint of reference is the parties’ corresponding objective intentions judgedfrom the observer’s perspective, in the context of their dealings andassuming honesty and reasonableness on all sides.

(iii) Contract ‘Void’ for Mistake

The importance of the formation/vitiation and term/fact distinctions isobscured by the description that contracts tainted by mistake are ‘void’ atcommon law. This word fails to distinguish between two categoricallydifferent situations. The first situation is one where no contract ever comesinto being because there was no corresponding offer and acceptance fromthe objective point of view.144 In the second situation, there is a prima facievalid contract that is not binding in the circumstances since it is defeasiblebecause vitiated for mistake of fact. The distinction between mistakeswhich ‘negative’ consent and those which ‘nullify’ consent, made by LordAitkin’s judgment in Bell v Lever Brothers Ltd, reaches towards the samedistinction but it is not sufficiently transparent to be helpful.145

C. Fundamental Common Mistaken Assumption

Since ‘mistake of terms’ is about contract formation, it is entirely logicalthat any objectively determined non-correspondence of the parties’ offerand acceptance should prevent contract formation (no contract resultsfrom my offering to sell new oats and you purporting to accept old oats).

142 Westminster Properties v Mudd, above n 123, Mendelssohn v Normand Ltd [1970] 1QB 177 (CA).

143 Lovell and Christmas, above n 65. Even the presence on an ‘entire agreement’ clause inthe contract does not prevent rectification, see JJ Huber, above n 65.

144 There may also be a contract but not on the terms sought to be enforced (because oneparty knew or had reason to know that the other’s intention did not correspond with thewords used and what the latter intended). See text accompanying nn 84–91.

145 Above n 141, at 217, cited, eg, in Peel, above n 4, at [8–001].




An attempted agreement has failed. On the other hand, contract partiesmust generally take the risk of mistaken assumptions outside that guaran-teed by the contract terms. This is necessary to protect the security oftransactions and contract parties’ reasonable expectations (that is theinstitution of contract). Hence, the contract stands if it is for new oats (orjust ‘oats’) but you merely believe (assume) they are old.

The correct classification shows that the real question is not why afundamentality requirement is absent for operative mistake of terms,146 butwhy relief should be given for mistakes of fact at all. The law shows thatthe legitimate concern to protect the security of contracts and the parties’reasonable expectations is outweighed when two conditions are satisfied.First, a claimant must show that his or her mistake was sufficiently serious(‘fundamental’) to divest their consent of significance in the actual (unmis-taken) circumstances. The mistake may, for example, make the contractpointless147 or unachievable.148 A claimant’s cry is not really that he or shedid not consent, rather, it is that he or she did consent, but not toperformance in these circumstances, the risks of which the claimant neitheranticipated nor expressly or impliedly assumed. The analogous doctrine offrustration discharges a contract from the occurrence of the frustratingevent because that is when a claimant’s consent ‘runs out’.

Second, the claimant must show that the other party shared this mistake.Pragmatically, this corroborates the claimant’s assertion of mistake (all tooeasy to make) and the importance of the mistake.149 More importantly, thisremoves the other party’s claim to the protection of his or her expectationsince that is also tainted by the catastrophic mistake. Voiding the contractin such circumstances deprives the other party of benefits (includingunexpected windfalls) which are unworthy of protection because the other

146 Peel, above n 4, at [8–044].147 For example, buying one’s own property as in Cooper v Phibbs (1867) LR 2 HL 149.148 See, eg, The Great Peace, above n 139. In that case, the question was whether the GP

hired by T was actually so far away from T’s distressed ship ‘at the time of the contract as todefeat the contractual purpose [to provide escort and standby services for five days until therescue tug arrived]—or in other words to turn it into something essentially different from thatfor which the parties had bargained? This is a question of fact and degree’: Great PeaceShipping Ltd v Tsavliris Salvage (International) Ltd (2001) 151 NLJ 1696 (QBD) [56](Toulson J)). The Court of Appeal concluded that although the GP was 400 miles (39 hourssailing) away, rather than 35 miles (three hours sailing) away as believed, it was not so faraway as to defeat the parties’ common assumption that it could still render the rescue servicedesired. This was reinforced by the T’s failure to cancel the agreement on discovering the truedistance between the ships until they found and hired a nearer vessel to assist. On the otherhand, Toulson J said that if there was five day’s sailing distance between the ships, thecontract would be void since its purpose would be unachievable: ibid, at [55].

149 Shared assumptions are more likely to relate to the essential substance of the contract,whilst unilateral mistakes are more likely relate to matters of importance only to the claimant(and not going to the substance of the contract).




party could not reasonably have expected them when he or she entered thecontract. The same analysis applies to contracts discharged for frustra-tion.150

D. Misrepresentation

What difference, if any, does it make if one party’s alleged ‘mistake’ isinduced by the other party’s statement? Recognition that misrepresenta-tions merely induce mistakes should be reflected in any coherent legalscheme. If the buyer’s ‘mistake’ in Smith v Hughes had been as to a term(that is he thought the seller was promising ‘old oats’), and if the mistakehad been induced by the seller’s statement, then the buyer should beentitled to treat the seller as having consented to sell old oats.151 The buyercan resist the seller’s claim for payment, or sue for non-delivery, of the oldoats. This is consistent with Curtis v Chemical Cleaning & Dyeing CoLtd152 and the remedy of rectification for common mistake in recordingthe parties’ agreement discussed above.153 Logically, the case for rectifica-tion should be stronger, not weaker, if one party’s misrepresentation, evenif innocent, induced the shared belief in the irregular meaning of wordsused in the document.154

We have noted that if the claimant enters a contract under a unilateralmistake as to the nature of the document or the identity of the other partyknown to the latter, or a shared mistake as to fact or assumption, thecontract will only be void if the mistake was ‘fundamental’. In contrast, amisrepresentation need not be fundamental to be actionable, but it onlyrenders the contract voidable;155 the right to rescission being vulnerable tovarious ‘bars’. Coherence demands that the claimant should be able to voidthe contract if his or her induced mistake about the nature of thecontractual document or the identity of the other party is fundamental.The contract should also be void if the claimant’s mistaken assumption isfundamental and induced by the other party’s innocent misrepresentation(so that the mistake can be characterised as common or shared). Noprinciple of subsidiarity should apply here.

150 See M Chen-Wishart, Contract Law, 2nd edn (Oxford, Oxford University Press, 2008)290–91.

151 See the analogous discussion in text accompanying nn 81–3, 136–8.152 Above n 70.153 See text accompanying nn 65–70.154 See to the contrary Rose v Pim, above n 31.155 Damages are also available for negligent, fraudulent and ‘Section 2(1) of the Misrep-

resentation Act 1969’ misrepresentations.




Misrepresentation should not trump or expel a concurrent action formistaken identity, non est factum156 or mistake in fundamental assump-tions. The claimant should not be worse off because his or her fundamentalmistake was induced by the defendant rather than being spontaneous.Conversely, the defendant should not be better off by inducing theclaimant’s mistake. In principle, this will make more contracts void whenthe concern to protect innocent third party purchasers pushes the otherway, in the direction of voidability. The latter concern is legitimate butshould be dealt with separately and transparently, rather than via twistingand manipulating contract law principles. As mentioned earlier, the tailshould not wag the dog.

IV. CONCLUSION

I began with just one troublesome case. Following its thread drew me deepinto the mire of the law of contractual mistake because, in a sense, Smith vHughes is symptomatic of that tangled area of the law. From this exegesis,three broad conclusions emerge. The first is that we need to take great careto define what we mean by the ‘objective’ or ‘subjective’ tests of intention.Needless muddle can arise when the label is used to designate quitedifferent points of reference. The second conclusion is that once thecontours of the objective test are properly understood, the cases hithertocharacterised as ‘mistake of terms’, including Smith v Hughes, are revealedas straightforward applications of objectivity. There is no need to resort tothe incoherent interpretation of exceptional subjectivity trumping theobjective approach. Indeed, such an explanation becomes nonsensical. Thedoctrine of rectification emerges as the natural corollary of this objectivetest of intentions. The third conclusion is that the attempt to understandwhy mistakes of term only have to be known by the other party (yetmistakes of fact must be shared and fundamental) in order to voidcontracts, requires us to stabilise the language used in the law of contrac-tual mistake. The looseness of the key descriptive and prescriptive terms—‘mistake’, ‘defective consent’ and ‘void’—allows quite different problemsto be thrown together causing inevitable confusion. Locating the primarydistinction between formation and vitiation allows more precise distinc-tions to be drawn: (a) ‘defective consent’ in the sense of ‘no objectiveagreement’ as opposed to ‘agreement but defeasible in the circumstances’;(b) ‘void’ in the sense of ‘no contract to start with’ as opposed to ‘prima

156 The majority in the Court of Appeal in Lloyds Bank plc v Waterhouse (1991) 10 Tr LR161 (CA) said that where the case is one of fraud or misrepresentation by the other party tothe contract, with no third party involved, the case should be dealt with as one ofmisrepresentation.




facie valid contract but set aside’; and (c) ‘mistake of terms’ in the sense of‘looking for correspondence in each party’s objective interpretation of theother’s intention regarding the obligations undertaken’, as opposed to‘mistake of fact’ as ‘mistaken assumptions not guaranteed by the contract’.

Identifying these points of reference helps to explain why knownnon-correspondence on any term prevents contract formation, while mis-taken assumptions (unless induced by misrepresentation) must be sharedand fundamental to void a contract. Moreover, we can begin to see howthe related areas of non est factum, mistaken identity and misrepresenta-tion should be classified and how troublesome issues under these headingsshould be resolved. The discussion can be summarised in the table below.Making sense of Smith v Hughes has been like untangling a giant knot. Allthe kinks are not straightened out but the picture looks promising.




TABLE: SUMMARY OF ANALYSIS

Formation: ‘Mistake of term’

(Whether and what agreement?)

Point of reference: Observer-contextualobjectivity except for non est factumapplying actor-contextual objectivity.

‘Void’: no contract formation or nocontract on the terms sought to beenforced.

Vitiation: Mistakenassumptions or ‘mistake offact’

(Whether void or voidablefor mistake?)

Point of reference:independently verifiablefact.

‘Void’: prima facie validcontract vitiated.

Observer-contextual objectivity

(i) Void if ‘known mistake of terms’: noobjective agreement on contract terms

(ii) Void if mistaken identity: noagreement to contract with the otherparty

(iii) Void if non est factum:fundamental unilateral mistake aboutthe nature of document (asactor-contextual objectivity) and actorunder disability and not careless

(iv) Void if incurable uncertainty

(v) Rectification of contractualdocument at equity if:

—mistake unilateral and other partymore blameworthy

—both parties agreed to differentwords or meaning

—misrepresentation

(i) Void if commonfundamental mistake

(ii) Voidable if ‘equitablemistake’, ie lessfundamental commonmistake but still serious andshared (doubtful after TheGreat Peace); or ifunilateral but the otherparty is more blameworthy

(iii) Voidable formisrepresentation




15

From Morgan to Etridge: Tracingthe (Dis)Integration of Undue

Influence in the United Kingdom

RICK BIGWOOD*

I. INTRODUCTION

‘THE LAW IN general leaves every man at liberty to make suchbargains as he pleases,’ wrote Sir John Salmond in Brusewitz vBrown,1 ‘and to dispose of his own property as he chooses.’2 He

continued:

However improvident, unreasonable, or unjust such bargains or dispositionsmay be, they are binding on every party to them unless he can prove affirma-tively the existence of one of the recognized invalidating circumstances, such asfraud or undue influence.3

Granted, undue influence has long been recognised as a substantial reasonfor setting aside an objectively concluded transaction, be it contract orgift,4 but what, when he wrote these words in 1922, did Salmond Junderstand ‘undue influence’ to mean?

* I am grateful to Joachim Dietrich, Graham Ferris, Pauline Ridge and Peter Watts fortheir comments and advice on this article. Thanks are also due to participants at the‘Exploring Contract Law’ symposium, University of Western Ontario, many of whomfurnished comments and support when the paper was first delivered. The usual caveatsapply.

1 [1923] NZLR 1106 (NZSC) 1109.2 Ibid.3 Ibid.4 I address inter vivos transactions only. I am not concerned with undue influence

affecting testamentary gifts, which expresses a different principle and is governed by theprobate court. For a recent discussion on the relationship between equitable undue influenceand undue influence in probate law, see P Ridge, ‘Equitable Undue Influence and Wills’(2004) 120 LQR 617.



A. Two Types of Undue Influence

At that time, and until 1985, the law relating to undue influence, thoughon occasion undoubtedly imperfectly expressed, was in the United King-dom, as in other major Commonwealth jurisdictions, both conceptuallytractable and (more or less) clear in terms of its criteria and purposes. Onebranch of the equitable jurisdiction—variously called ‘actual’, ‘express’,‘non-relational’ or ‘Class 1’ undue influence—regulated conduct that wasstraightforwardly coercive, exploitative, manipulative or deceptive towarda peculiarly vulnerable party. Such conduct was unacceptable for thatreason simpliciter. Mostly this was equity’s counterpart to the commonlaw’s erstwhile, overly circumscribed duress doctrine, and it was closelyrelated to, if not merely a manifestation of, equity’s cognate jurisdiction torelieve against equitable fraud in the manner of ‘unconscionable’ (or‘unconscientious’) dealing.5 Thus, characteristically, the Class 1 undueinfluence jurisdiction regulated objectionable conduct affecting another’swill, hence that was ‘responsibility-relieving’ from the victim’s standpoint,even if it occurred outside a fiduciary relationship. The category was,therefore, distinctively non-relational, or ‘relationship-independent’, in thesense that it was not functionally directed at preserving the integrity of ageneric class of task or relation, except to the extent that such a task orrelation might have, in the particular instance, afforded an opportunity for‘actual’ or ‘overt’ wrongdoing (for example coercion) to actualise betweenthe parties concerned.6 The focus of Class 1 undue influence was, accord-ingly, purely transactional and remedial, the public basis of interventionbeing the principle that ‘no one shall be allowed to retain any benefit fromhis own fraud or wrongful act’.7

The other branch of the jurisdiction, in contrast—variously called‘presumed’, ‘relational’ or ‘Class 2’ undue influence—was of much nar-rower scope and sui generis. It was, distinctively, a contextual applicationof fiduciary accountability. ‘Undue influence’ here possessed a ratherdifferent meaning than under the Class 1 jurisdiction. It involved a form ofcomplaint that could only occur within a fiduciary relationship—hence itslong-standing association with relations involving ‘dependence and trust’,

5 Earl of Chesterfield v Janssen (1751) 2 Ves Sen 125, 28 ER 82 (Ch) 155–6 (LordHardwicke). The phrase ‘undue influence’ appears to have been first used by Lord Hardwickein Morris v Burroughs (1737) 1 Atk 399, 26 ER 253 (Ch) 403.

6 As occurred, for instance, in Re Craig (1970) [1971] 1 Ch 95. Cf Royal Bank ofScotland plc v Etridge (No 2) [Etridge] (2001) [2002] 2 AC 773 (HL) [103] (Lord Hobhouse).

7 Allcard v Skinner (1887) 36 Ch D 145 (CA) 171 (Cotton LJ). Cf CIBC Mortgages Plcv Pitt [1994] 1 AC 200 (HL) 209 (Lord Browne-Wilkinson), explaining that the effect of thedefendant’s wrongdoing is to prevent the plaintiff from freely and informedly consenting tothe transaction, which ‘accordingly must be set aside in equity as a matter of justice’.

380 Rick Bigwood



‘trust and confidence’, and the like.8 Despite evolving and being adminis-tered as a stand-alone body of doctrine,9 Class 2 undue influence neverthe-less shared with wider fiduciary law the same generic functional purpose orsubstantive policy goal of controlling the narrow mischief of disloyalopportunism by those legally required to act for, on behalf of, or in theinterests of another.10 Although the jurisdiction shared with other exculpa-tory doctrines of the common law and equity a concern for ‘fair dealing’and the quality of interpersonal transactional consent—so that the law’sfunction in this field could in part be seen, too narrowly, as a merecorrective for one party’s ‘will’ having been wrongfully affected by anotherin the formation of the impugned transaction11—the additional andoverarching concern of the Class 2 undue influence jurisdiction wasundeniably to prevent abuses of trust, as understood in conventionalfiduciary terms.12 ‘Wrongfully influencing the claimant’s will’ was merelythe mechanism by which value was diverted in a manner inconsistent withthe purposes for which the parties’ special relation of influence existed, orin equitable contemplation was deemed to exist.13 The foundation forintervention in such cases was thus framed squarely by reference to ‘publicpolicy’ (‘trust maintenance’, ‘risk management’, ‘prophylaxis’) rather thanin terms of a specific ‘wrongful act’ having been proved against theinfluential party on the normal civil preponderance.14 It followed that

8 The relations that attract the jurisdiction have been variously described in thejurisprudence on the subject. Some courts have observed how the expressions ‘relation ofinfluence’, ‘relation of confidence’, and ‘fiduciary relation’ are often used interchangeably inthis context, while reminding the reader that they are not necessarily coextensive inapplication: see Jenyns v Public Curator (Q) (1953) 90 CLR 113 (HCA) 132 (Dixon CJ,McTiernan J and Kitto J). As will become clear below, special influence is legally presumed ina few well-known types of relation (eg solicitor–client, parent–child, doctor–patient, andguardian–ward). These are generally known today as ‘Class 2(A)’ relationships of influence.However, special influence can also exist outside of these traditional, status-based relations,and the plaintiff may prove it as a fact in the circumstances of the particular case. The label‘Class 2(B)’ undue influence is commonly assigned to such cases. It is vital to bear in mindthat the type of relation in question is identical as between the two sub-classes of Class 2undue influence. The nature of the special relation or function that attracts the Class 2jurisdiction is discussed below at text accompanying nn 72–9.

9 Paul Finn claims that this is ‘more for historical reasons than for reasons of soundprinciple’: PD Finn, ‘The Fiduciary Principle’ in TG Youdan (ed), Equity, Fiduciaries andTrusts (Toronto, Carswell, 1989) 43.

10 Here meaning the unauthorised diversion of the value of assets falling within the scopeof the particular fiduciary function or task.

11 See, eg, Commercial Bank of Australia Ltd v Amadio (1983) 151 CLR 447 (HCA)461, where Mason J explained that relief for undue influence is given because ‘the will of theinnocent party is not independent and voluntary because it is overborne’.

12 See, eg, LA Sheridan, Fraud in Equity: a study in English and Irish law (London,Pitman, 1957) ch 5. See also Part II below.

13 ‘[A]ffecting another’s “will” is merely one way in which an advantage can be taken ina relationship of ascendancy or trust’: Finn, above n 9, at 45.

14 The clearest statement of this is found in Cotton LJ’s judgment in Allcard v Skinner,above n 7, at 171: ‘In the [Class 2 undue influence] cases the Court interferes, not on the

From Morgan to Etridge 381



undue influence could be presumed, and in several cases was presumed,15

even in the absence of evidence of any actual wrongdoing or exercise ofpower (that is, beyond that which could be inferred provisionally from thenature of the relation and the particular transaction inter se).16 Publicpolicy dictated that relations or tasks marked by the fiduciary functionshould be protected from the mere possibility of abuse, including abuse inthe manner of an unauthorised use of any special capacity to influence thatwas associated with such relations or tasks.

In practical terms the protection in the Class 2 line of cases took theform of particularised procedural rules that were designed to safeguard theintegrity of fiduciary relationships, tasks or functions by managing the risk,when real,17 against natural human tendencies to disloyalty. The most

ground that any wrongful act has in fact been committed by the donee, but on the ground ofpublic policy, and to prevent the relations which existed between the parties and the influencearising therefrom being abused.’ In my view this passage is apt to confuse, and requiresexplanation. Public policy justifies giving artificial effect to the evidence via a presumption ofundue influence, but it is whatever is presumed thereby (undue influence) that justifies courtinterference, even though, at the same moment, the court must accept that undue influencemay not have been proved in fact on the normal civil preponderance. In other words, publicpolicy is the source of the presumption, but not its content. The ‘presumption’ is not thatundue influence might have occurred (which is the reason for the presumption), but ratherthat it did occur (which is the reason for exculpation from the transaction) (see also n 18below). It follows that once the presumption is properly activated, the evidence in reply mustaddress the content of the presumption and not the (public policy) reasons for thepresumption triggering in the first place. The effect of a non-displaced presumption is that, inthe absence of some defence being available to a defendant, the presumption is controllingand the transaction will be set aside on the ground of ‘undue influence’ rather than ‘publicpolicy’ per se.

15 See, eg, Allcard v Skinner, above n 7, at 183 (Lindley LJ); Johnson v Buttress (1936) 56CLR 113 (HCA) 133–4 (Dixon J), 143 (McTiernan J); Hartigan v International Society forKrishna Consciousness Inc [2002] NSWSC 810.

16 This is not to be confused with the suggestion that a transaction can be impeached onthe ground of relational undue influence even if the defendant’s conduct is found to be‘unimpeachable’: cf Hammond v Osborn [2002] EWCA Civ 885, [32] (Nourse LJ), [61](Ward LJ). If the defendant is found not to have exercised undue influence, hence actedconflictually (that is, self-interestedly or inconsistently with his or her special duty of loyalty),it cannot follow that he or she can be presumed, counterfactually, to have done so: cf Geffenv Goodman Estate (1991) 81 DLR (4th) 211, [1991] SCR 353, 244 (Sopinka J). As LordScott pointed out in Royal Bank of Scotland plc v Etridge (No 2), above n 6, at [219]: ‘itmakes no sense to find, on the one hand, that there was no undue influence but, on the otherhand, that the presumption applies,’ although his Lordship’s comments are made on the basisthat ‘presumed’ undue influence is merely a form of ‘actual’ undue influence proved by way ofa non-contradicted permissible inference.

17 As I have discussed elsewhere, there must be ‘a conflict or a real or substantialpossibility of a conflict’ in order to trigger the conflict rule/presumption of undue influence: RBigwood, ‘Undue Influence in the House of Lords: Principles and Proof’ (2002) 65 MLR 435,447–9. The rule/presumption will not be triggered if the influential fiduciary’s opportunity topursue a personal (or a third party’s) interest (a) though in itself strong enough to be aninducement, is too remote from the actual area of the fiduciary’s responsibilities vis-à-vis hisbeneficiary, or (b) though itself sufficiently proximate to those responsibilities, is too weak aninducement realistically to be a possible determining motive on the part of the fiduciary. Thesubordinate party thus has no prima facie complaint, and the fiduciary will not be called on

382 Rick Bigwood



significant of those rules involved the ritualistic implementation of apresumption of undue influence that effectively authorised a court to giveartificial effect to the claimant’s evidence. Being informed by the conven-tional fiduciary rationale, such a presumption was motivated by the strongpublic desire to avoid unauthorised conflicts of interest and duty by thosewho were required to act exclusively in the interests of another. Hence, assoon as the right combination of circumstances existed—opportunity(fiduciary influence), incentive (the impugned benefit) and epistemologicaluncertainty (serious detection and evidentiary problems)—equity indulgedin a presumption that self-interest and temptation had operated in theparticular case: that what was feared had materialised.18 It presumed thatfiduciary influence had been exercised, actively or passively, and in anyevent conflictually,19 in the procurement or receipt of the impugnedbenefit, and it cast upon the benefiting party, if he or she wished tomaintain the benefit, the burden of positively demonstrating the righteous-ness of the transfer or conveyance20—of showing that undue influence hadnot been exercised, but rather that the benefit received was ‘the independ-ent and well-understood act of [the plaintiff]’, who had been ‘in a positionto exercise a free judgment based on information as full as [the defendant’sown]’.21 Put another way, ‘The risk is that [the defendant] may put his owninterests ahead of those of [the plaintiff]. The obligation is to prove that hehas not done so.’22 As in regular fiduciary contexts, this shifting of the

to justify or explain, where the personal opportunity taken up is, realistically, unconnected orunrelated to his or her fiduciary office or function (eg, in the negotiation of a professional feefor the fiduciary’s services), or where, again realistically, the beneficiary has no demonstrableinterest or expectancy in the subject matter of the transaction inter se (eg, trivial or moderategifts). Indeed, such exclusions from the ambit of the conflict rule are merely acknowledge-ments of the purpose of the rule itself: as opportunity and temptation abate, so too doesincentive, and so too does the risk of abuse (fiduciary opportunism). The influential fiduciaryis accordingly free to take up the opportunity as if the relationship were an arm’s-length one.

18 As Salmond J made clear in Brusewitz v Brown, above n 1, at 1110: ‘[A relation inwhich undue influence occurs is] … such a relation of superiority on the one side andinferiority on the other … and therefore such an opportunity and temptation for theunconscientious abuse of the power and influence so possessed by the superior party, as tojustify the legal presumption that such an abuse actually took place and that the transactionwas procured thereby’ (emphasis added). Instructive in this connection, too, is JC Shepherd,The Law of Fiduciaries (Toronto, Carswell, 1981) 149–50.

19 That is to say, it was presumed that the impugned benefit was the result of thedefendant acting inconsistently with the responsibility to use his or her special influentialcapacity exclusively in the interests of the plaintiff, who was subject to the influence and hadentrusted the defendant with his or her welfare.

20 Note that the burden was on the defendant to show affirmatively that the transactionwas the plaintiff’s free and properly understood act, or otherwise not the result of an abuse ofhis or her position. This meant that the plaintiff’s independence could not be shown simply bythe absence of any evidence that the defendant had exercised influence over the plaintiff.Generally see Johnson v Buttress, above n 15, discussed below at text below accompanyingnn 84–95.

21 Johnson v Buttress, ibid, at 134 (Dixon J).22 Re P’s Bill of Costs (1982) 8 Fam LR 489, 496 (Evatt CJ and Fogarty J).




burden of proof onto the defendant to disprove undue influence—todemonstrate that there had in fact been no unauthorised conflict of interestor duty that operated to vitiate transactional consent—served to protectfiduciary relationships, tasks or functions by helping to ensure that thoseentrusted with the fiduciary function acted consistently with the responsi-bilities that such a function entails. Given the inherent difficulty of provinga negative to this effect, the application of the presumption effectivelyimposed strict liability, even if in theory liability for undue influence wasbased on ‘fault’ or ‘blame’.23 Again, this served the substantive goals offiduciary law by holding those entrusted with fiduciary tasks to a higher(that is, a strict) ethic, thereby removing any possible incentive to disloy-alty. It also reflected the greater importance of ‘prevention’ as the goal offiduciary rules relative to the wider civil law (tort, contract, equitablewrongs such as Class 1 undue influence, estoppel and unconscionabledealing), where the paradigm is ‘repair’ for the doing of harm rather thanthe prevention of harm.24

The presumption of undue influence was, therefore, a pragmatic legalconstruction rather than ‘genuinely evidential’. It was an exercise inapplying, mutatis mutandis, the conventional ‘no-conflict’ rule to theparties’ specific relationship and transactional encounter. It was a tech-nique by which the law expressed and effectuated a value judgment aboutthe strength of the various interests to which ‘responsibility judgments’relate in the fiduciary context.25 It was, like other policy-based presump-tions, a self-conscious judicial response to societal pressures and values inconnection with a particularly invidious mischief, one that transcended thesimple rational relationship between the basic evidentiary facts and thepresumed fact in the individual case. The evidential effect of the presump-tion was thus significantly greater than the natural probative weight orinferential worth of the basic facts that sufficed to generate the presump-tion, were those facts assessed in isolation from the substantive policyreasons for the presumption. The basic facts merely had to give rise to a‘realistic suspicion’ of undue influence, as distinct from a ‘legitimateinference’ of it. Again, once the circumstances of actual or possible conflictbetween personal interest and fiduciary duty were present, equity inter-vened to prevent the mere possibility of self-interest being preferred toduty, in disputes where, moreover, the circumstances typically renderedproof of actual wrongdoing impossible, or at least extremely difficult.26

23 Cf P Cane, Responsibility in Law and Morality (Oxford, Hart Publishing, 2002) 46(concerning res ipsa loquitur), 91–2.

24 Cf Cane, ibid, at 133.25 I paraphrase here a point made by Cane, ibid, at 91.26 Cf Re Craig, above n 6, at 104 (Ungoed-Thomas J). Such difficulty of proof exists

typically because of the secret nature of the dealings between the plaintiff and the defendant,and often also because of the death of the transferor.

384 Rick Bigwood



And although this increased the risk of misattribution of liability, and sopossibly worked a hardship upon factually innocent fiduciaries in somecases, this was a risk worth tolerating given the wider interest that societyhas in maintaining fiduciary relations or tasks.27 In other words, thepresumption of undue influence was a significant concession to an overtpublic policy. It was intended to benefit claimants in a material way, bymaking it easier for them to succeed in their claims against influentialfiduciaries and by making defences harder to maintain.28 Even though thebasic facts that triggered the presumption may not have possessed thelogical value to justify a finding (via inference) of undue influence as matterof probability, public policy nevertheless validated the courts in bridgingthe gap and giving the logical connection between the basic facts and thepresumed fact an added evidential weight that it would not ordinarilycarry. Again, this was a deliberately constructed attempt, especially on thepart of nineteenth-century equity judges, to manage against the perniciousrisk of conflicts of interest and duty in fiduciary settings.29

It followed from all this that relational undue influence was ‘wrongful’because, unlike what was occurring in the arm’s-length Class 1 cases, andwhatever else was actually involved in the individual case (coercion,misrepresentation, incapacity or the like), it entailed conflictual conductand hence liability: it involved the unauthorised use, or presumed unau-thorised use, of influence in a relationship where self-denial was both

27 As Lamond J observed in Bradley v Crittenden [1932] SCR 552, [1932] 3 DLR 193,569: ‘The rule of equity which places on the donee the burden of proving both the gift and theindependence of the donor’s will in making it, may be a harsh one and, in individual cases,may lead to hardship. The courts, however, have found it necessary to maintain it in order toprevent those in a position to exercise undue influence from taking advantage if their positionunder circumstances in which proof thereof would be impossible.’

28 In other words, although on one level the presumption could be seen merely to manageforensic difficulties, it was more than that. It incorporated a substantive judgment that thosetransferring suspicious benefits within the scope of a fiduciary relationship should succeed intheir claim against the fiduciary if there was room for doubt as to the motives that inspiredthe transfer in the particular case. The risk of non-persuasion on the issue of relational undueinfluence shifts when the realistic possibility of such wrongdoing is shown to exist. Whenaided by the presumption, the plaintiff wins summarily (if the defendant cannot rebut thepresumption) even though the evidence the plaintiff supplied under the burden of productionwould not otherwise have satisfied the burden of persuasion. The presumption assists theplaintiff in a material way, by treating him or her as if he or she had met the persuasiveburden, which is purely a concession to the substantive policy that informs this area of thelaw.

29 Lord Eldon put it very strongly in two judgments: in Gibson v Jeyes (1801) 6 Ves Jun266 (Ch) 276, 31 ER 1044 he stated: ‘It is necessary to say broadly, that those, who meddlewith such transactions, take upon themselves the whole proof, that the thing is righteous. Thecircumstances, that pass upon such transactions, may be consistent with honest intentions:but they are so delicate in their nature, that parties must not complain of being called on toprove, they are so,’ and in Hatch v Hatch (1804) 9 Ves Jun 292 (Ch) 297, 32 ER 615 hestated: ‘[I]f the Court does not watch these transactions with a jealousy almost invincible, ina great majority of cases it will lend its assistance to fraud.’




expected and required on the part of the influential actor. This also meantthat it was, strictly speaking, pointless to inquire into the precise manner ofthe exercise of the special influence in such cases: it was unnecessary to askwhat was being ‘presumed’ in a successful Class 2 undue influence claim.Regardless of how fiduciary influence is exercised in relational undueinfluence situations (pressure, flattery, advice, argument, pleading, interces-sion, non-disclosure, concealment of self-interest, discouraging recourse toothers for advice and so on), the fact that such influence is successfullyemployed at all in a manner inconsistent with the defined and limitedpurposes of the parties’ special relation affords (all else being equal) asufficient juristic reason to suppress or reverse the impugned transaction asagainst the conflicted party. The general goal of fiduciary regulation is tocontrol opportunism in limited-access arrangements, whatever form ittakes. Relational undue influence, therefore, never regulated a discrete andwell-defined modus operandi of victimisation like the law relating to otherforms of objectionable conduct such as duress and misrepresentation, andso it has not lent itself to any concrete definitional approach in the mannerthat most other exculpatory categories have allowed. To be sure, all that is‘presumed’ when the ‘presumption’ of undue influence operates is that thedefendant has, through his or her special influential capacity over theplaintiff, and without consent or authorisation, preferred self-interest toduty, and that the plaintiff has, as a result, become the victim of misplacedtrust or confidence vitiating the transactional consent.

B. The Law Changes

In 1985, this reasonably clear and stable picture of the dual strands ofequity’s bifurcated jurisdiction to interfere with transactions on the groundof undue influence began to dissolve, at least in the United Kingdom. LordScarman, on behalf of the House of Lords in National Westminster Bankplc v Morgan,30 denied that the basis for relief in this area was ‘a vague“public policy”’; it was, in contrast, ‘specifically the victimisation of oneparty by the other’.31 Accordingly, in his Lordship’s view, no

presumption of undue influence [could] arise from the evidence of the relation-ship of the parties without also evidence that the transaction itself was wrongfulin that it constituted an advantage taken of the person subjected to the influencewhich, failing proof to the contrary, was explicable only on the basis that undueinfluence had been exercised to procure it.32

30 [1985] AC 686 (HL) [Morgan].31 Ibid, at 705, citing Lindley LJ in Allcard v Skinner, above n 7, at 182–3.32 Ibid, at 704.

386 Rick Bigwood



‘Presumed undue influence’ thus involved a presumption triggered by theprobability (or serious likelihood) of abuse rather than the mere possibilityof it. In relation to contractual transactions at least, this meant that thepresumption of victimisation in the manner of relational (Class 2) undueinfluence could not arise unless the impugned transaction was shown to be‘manifestly disadvantageous’ to the claimant, as judged by the objectivelyunfair terms of the impugned contract itself33—a stance very different fromthe classical fiduciary approach based on the ‘realistic possibility’ of abusethat had been applied by the Court of Appeal below.34

In 1994, obiter dicta by Lord Browne-Wilkinson in CIBC Mortgages plcv Pitt signaled that the traditional account of undue influence mighteventually be restored.35 His Lordship remarked that the House of Lordsmight need to consider ‘the exact limits of Morgan … in the future,’ and inparticular the relationship, if any, between the fiduciary principle andrelational undue influence.36 However, any hope of rigorous examinationof that relationship was frustrated when, in 2001, the House of Lords inRoyal Bank of Scotland plc v Etridge (No 2)37 failed to address (let alonerestore) the fiduciary account of relational undue influence. On thecontrary, and with virtually no reference to the prior decisions andacademic literature on the subject, the House of Lords asserted that‘presumed undue influence’ was not a distinct legal phenomenon, butrather merely an example of the way in which undue influence, typically inthe manner of ‘unfair persuasion’, could be proven in a certain category ofcase involving relationships that were peculiarly vulnerable to abuse byway of non-overt acts of persuasion. Undue influence, therefore, waspresented as an integrated and unitary legal phenomenon or doctrine, withthe effect that it now has a universal connotation regardless of therelational context in which it occurs. In all cases the claimant bears theonus of proving undue influence in fact, so that all successful claims in thisarea must necessarily involve undue influence that is ‘actual’ in the sensethat the claimant has made out his or her case against the wrongdoer onthe normal civil standard.38 However, in some cases—in particular thosefalling under the traditional nomenclature of ‘presumed’ undue influence—

33 Ibid.34 As Slade LJ in the Court of Appeal in National Westminster Bank plc v Morgan [1983]

3 All ER 85 (CA) 92 recognised, ‘Where a transaction has been entered into between twoparties who stand in the relevant relationship to one another, it is still possible that therelationship and influence arising therefrom has been abused, even though the transaction is,on the face of it, one which, in commercial terms, provides reasonably equal benefits for bothparties.’ See also Dunn LJ, ibid, at 90.

35 Above n 7 [Pitt].36 Ibid, at 209.37 Above n 6.38 See especially Lord Clyde, ibid, at [93].




the plaintiff succeeds by way of factual inference from naturally prepon-derating primary facts that were proven and insufficiently answered bycounter-evidence on the other side. In other words, ‘presumption’ heremeans ‘permissible inference’, and the mandatory effect and the prophylac-tic content and function of the traditional presumption are gone. Byimplication, the purpose of the jurisdiction is thus purely remedial, beingconcerned only with responding to the genuineness of the plaintiff’stransactional consent relative to the defendant’s ‘conscience’ in the particu-lar case, and it does not serve any higher regulatory function external tothe individual exculpatory claim itself (such as prophylaxis in relation tofiduciaries generically). Also, the principle in the so-called ‘relationship’cases was said in Etridge not to be confined to instances of ‘abuse of trustand confidence’; it extends as well to arm’s-length transactions—to ‘caseswhere a vulnerable person has been exploited’.39

The effects of Etridge are obvious if not entirely explicit in thejudgments. The fiduciary rationale is eviscerated, despite, as we shall see,their Lordships describing certain relationships that attract the jurisdictionin classic ‘fiduciary’ terminology. This explains the reference in this article’stitle to the ‘disintegration’ of undue influence in the United Kingdom: thatClass 2 ‘relational’ undue influence has been detached from its historical‘public policy’ foundations and no longer has special regulative effect. Butthis process of disintegration has been one of ‘integration’ as well. At leastfrom a forensic standpoint, Class 1 and Class 2 undue influence nowpossess a unitary, but still unavoidably vague, meaning. Moreover, undueinfluence is aligned both procedurally and (more unfortunately it seems)substantively with other exculpatory categories known to the common lawand equity such as duress, misrepresentation and unconscionable dealing(at least in its Antipodean form).40 So profound has this process ofintegration been in the United Kingdom that it has, in my view, becomealmost impossible to distinguish undue influence practically and intellectu-ally from certain other exculpatory categories—unconscionable dealingespecially41—such that it has lost the narrow meaning and specific function

39 Ibid, at [11] (Lord Nicholls).40 All references herein to ‘unconscionable dealing’ are to the developed antipodean

version of the doctrine, as expounded, eg, in Commercial Bank of Australia Ltd v Amadio,above n 11. For an extended discussion of that jurisdiction, see R Bigwood, ExploitativeContracts (Oxford, Oxford University Press, 2003) ch 6.

41 The similarities are remarkable, especially if one accepts the ‘shifting onus’ approachsometimes found in judicial formulations of the unconscionable dealing doctrine, such aswhen judges describe an equitable ‘presumption’ arising from proof of ‘inequality ofbargaining power’ and ‘substantial unfairness of the bargain obtained by the stronger party’.See, eg, Morrison v Coast Finance Ltd (1965) 55 DLR (2d) 710 (BCCA) 713 (Davey J); Harryv Kreutziger (1978) 95 DLR (3d) 231 (BCCA) 237 (McIntyre J); Commercial Bank ofAustralia Ltd v Amadio, above n 11, at 474 (Deane J). It is fairly clear, though, that the legalburden of proving unconscionable dealing rests throughout on the party asserting the claim,

388 Rick Bigwood



that it once possessed. Undue influence, therefore, risks becoming some-thing of an equitable ‘catch-all’ category, much in the way that its sibling,unconscionable dealing, seems increasingly to be utilised in Australia.42 Tomy mind it has become virtually impossible to understand the post-Etridgedoctrine of undue influence in any meaningful way, so as to identify, andmore importantly justify, its independent status apart from those othercategories, especially unconscionable dealing.

In this article I want to describe, rather than to fully defend, the classicalfiduciary account of relational undue influence. That account, I shallconclude, is defensible, but I am content for present purposes to treat it asa historical legal fact that subsisted in the United Kingdom until around1985, and which continues to subsist elsewhere today (for example inCanada and Australia). I want then to trace, relative to that account, theprogressive demise of relational undue influence as a category within theframework of fiduciary regulation, through the House of Lords decisionsin Morgan and Etridge especially, and to comment on the bases forrejection of the fiduciary rationale. In Morgan the bases for rejection ofthat rationale were not developed in a convincing way, and in Etridge,where one might have expected concentration on the subject in the light ofLord Browne-Wilkinson’s obiter dicta in Pitt, assertion appears to havesubstituted for precedent and analysis.

II. ELABORATING RELATIONAL UNDUE INFLUENCE AS A FIDUCIARYRULE43

From a legal history standpoint, the essential fiduciary nature of relationalundue influence is undeniable. It can be traced through a respectable line ofnineteenth and twentieth-century authorities,44 in addition to Dixon J’s

and that ‘presumption’ here means ‘permissible inference’, with a shift in the evidential onusonly: see, eg, Diprose v Louth (1992) 175 CLR 621 (HCA) 632; Micarone v PerpetualTrustees Australia Ltd (1999) 75 SASR 1 (SC) 127; Smyth v Szep [1992] WWR 673 (BCCA)681–2; Gindis v Brisbourne (2000) 183 DLR (4th) 431 (BCCA) 442–3.

42 ANZ Banking Group Ltd v Karam (2005) 64 NSWLR 149 (CA) where the courtrecommended absorption of lawful-act duress into unconscionable dealing. It might be arguedthat unconscionable dealing is better suited to regulating unfair persuasion in arm’s-lengthtransactions than relational undue influence.

43 Material in some subsections of this section has been reproduced, though mostly in anedited form, from Bigwood, above n 40, ch 8.

44 See, eg, Gibson v Jeyes, above n 29, at 1049–50; Huguenin v Baseley (1807) 14 VesJun 273, 33 ER 526 (Ch) 531–532; Billage v Southee (1852) 9 Hare 534 (Ch) 540, 68 ER623; Wright v Vanderplank (1856) 8 De GM & G 133 (Ch) 137, 44 ER 340; Tate vWilliamson (1866) 2 Ch App 55, 60; Allcard v Skinner, above n 7, at 171; Bradley vCrittenden, above n 27, at 559; Tufton v Sperni [1952] 2 TLR 516 (CA) 522 (Sir RaymondEvershed MR), 530 (Jenkins LJ); Lloyds Bank Ltd v Bundy [1975] QB 326, especially thejudgment of Sir Eric Sachs; O’Sullivan v Management Agency and Music Ltd [1985] QB 428,




exemplary account of the subject in Johnson v Buttress,45 which will bediscussed further.46 There was, and indeed still is, significant academicopinion in support of the fiduciary personality of relational undue influ-ence in English-based legal systems, including the United Kingdom, Aus-tralia, Canada and the United States.47 Understanding undue influence as acategory within the framework of fiduciary regulation, however, can bemade difficult by reliance on apparent, and often non-conceded or unap-preciated, misunderstandings and misdescriptions of the conventionalboundaries of fiduciary accountability by modern courts and commenta-tors.48 Indeed, as will be seen below, virtually all denials of the essentialfiduciary nature of relational undue influence law involve a basic failure toexplore, appreciate or respect the conventional boundaries of fiduciaryobligation and regulation. Below I set out my own understanding of thoseboundaries, with a view to showing how relational undue influence restsnaturally within them.49

448–9. For an explicit acknowledgement, after Morgan, of the substantial identity betweenundue influence and breach of fiduciary duty, see Goldsworthy v Brickell [1987] Ch 378,400–401 (Nourse LJ).

45 Above n 15.46 See text below accompanying nn 84–95.47 Examples include S Williston, Williston on Contracts (New York, Baker, Voorhis,

1937) § 1625ff; Sheridan, above n 12, at ch 5; LS Sealy, ‘Fiduciary Relationships’ [1962] CLJ69, 78; PD Finn, Fiduciary Obligations (Sydney, Law Book Company Ltd, 1977) 41ff;Shepherd, above n 18, at ch 14; J Glover, Commercial Equity: Fiduciary Relationships(Sydney, Butterworths, 1995) 8ff, 33; AJ Duggan, ‘Undue Influence’ in P Parkinson (ed), ThePrinciples of Equity, 2nd edn (Sydney, Law Book Company Ltd, 2003) 428–31; R Flannigan,‘The Fiduciary Obligation’ (1989) 9 OJLS 285; PD Maddaugh and JD McCamus, The Lawof Restitution, 2nd edn (Toronto, Canada Law Book, 2004) 812; G Ferris, ‘Why is the Law ofUndue Influence so Hard to Understand and Apply?’ in E Cooke (ed), Modern Studies inProperty Law—Volume 4 (Oxford, Hart Publishing, 2007) ch 3. Ashburner viewed thepresumption of undue influence, as a ground for rescinding contracts, as ‘so closely connectedwith the doctrines [relating to conflict of interest and duty]’ that he dealt with them in aseparate chapter dedicated to that subject: see D Browne, Ashburner’s Principles of Equity,2nd edn (London, Butterworths, 1933) 299, ch 21.

48 As to which see generally R Flannigan, ‘The Boundaries of Fiduciary Accountability’(2004) 83 Canadian Bar Review 35, [2004] New Zealand Law Review 215. The referencesherein are to the New Zealand Law Review reprint of the article.

49 As the following text and footnotes reveal, my understanding of the conventionalboundaries of fiduciary accountability owes a large debt to Robert Flannigan’s lucid andsustained writings in the field. These include Flannigan, above n 47; R Flannigan, ‘Commer-cial Fiduciary Obligation’ (1998) 36 Alberta Law Review 905; R Flannigan, ‘FiduciaryRegulation of Sexual Exploitation’ (2000) 79 Canadian Bar Review 301; Flannigan, above n48; R Flannigan, ‘Fiduciary Duties of Shareholders and Directors’ [2004] Journal of BusinessLaw 277; R Flannigan, ‘A Romantic Conception of Fiduciary Obligation’ (2005) 84Canadian Bar Review 391; R Flannigan, ‘The Adulteration of Fiduciary Doctrine inCorporate Law’ (2006) 122 LQR 449; R Flannigan, ‘The Strict Character of FiduciaryLiability’ [2006] New Zealand Law Review 209; R Flannigan, ‘The Economics of FiduciaryAccountability’ (2007) 32 Delaware Journal of Corporate Law 393; R Flannigan, ‘The[Fiduciary] Duty of Fidelity’ (2008) 124 LQR 274.

390 Rick Bigwood



A. Fiduciary Regulation in a Nutshell

Fiduciary regulation exists exclusively to ensure selfless loyalty by thosewho are required to act for, on behalf of, or in the interests of, another. Itserves no other juridical purpose.50 Although a fiduciary obligation isimposed51 on certain arrangements by private law, the basic justificationfor recognising and regulating relationships or tasks of a fiduciary nature isessentially a ‘public’ or ‘social’ one.52 It is ‘informed by considerations ofpublic policy aimed at preserving the integrity and utility of [fiduciary]relationships [or tasks] given the expectation that the community isconsidered to have of behaviour in them, and given the purposes they servein society’.53 It is also justified by the significant opportunities that exist forfiduciaries as a class to act inconsistently with the equitable purposes oftheir office or function (the risk of the mischief sought to be suppressed bythe imposition of fiduciary obligation is serious and high) in combinationwith the monitoring and evidentiary problems that characteristically attend

50 Nominate duties on the part of fiduciaries that go beyond exacting loyalty (eg, dutiesof care), or defalcations by fiduciaries that do not involve disloyalty per se (eg, discretionaryunequal distributions to beneficiaries) are not fiduciary duties or defalcations, and hence donot attract fiduciary accountability or liability. According to Flannigan, fiduciary accountabil-ity ‘is concerned exclusively with controlling opportunism on the part of those with limitedaccess. It is not concerned with defining substantive performance standards for nominatefunctions. Nor is it concerned with lack of care, bad judgment, unjust enrichment, powerdifferentials, general market exploitation or the substantive merits of decisions’: R Flannigan,‘Fiduciary Duties of Shareholders and Directors,’ above n 49, at 279. See also S Worthington,‘Fiduciaries: When Is Self-Denial Obligatory?’ [1999] CLJ 500, 501–503. We must thereforedistinguish between fiduciary obligations and the full range of idiosyncratic or nominateresponsibilities that may be owed by a fiduciary to his or her obligee. For judicial recognitionof this point, see Pilmer v Duke Group Ltd (in liq) (2001) 207 CLR 165 (HCA) 196, 217(Kirby J), citing Norberg v Wynrib (1992) 92 DLR (4th) 449, [1992] 2 SCR 226, 272(McLachlin J); Bristol and West Building Society v Mothew (1996) [1998] Ch 1, 16 (MillettLJ); Henderson v Merrett Syndicates Ltd (1994) [1995] 2 AC 145 (HL) 206 (LordBrowne-Wilkinson).

51 At root, fiduciary status and thus accountability is an ‘imposed’ form of obligationrather than an ‘assumed’ one. Although the physical arrangement that led to one party havingdefined and limited access to another party’s assets may have been the result of a voluntarydecision on the part of the parties involved (eg via contract or unilateral declaration of trust),it is the fact of the resultant limited access, and the risk of opportunism that flows from thataccess, rather than the consent to the physical arrangement itself that leads to the impositionof fiduciary obligation within the scope of that arrangement. Cf Finn, above n 9, at 46–7, 54;R Flannigan, ‘Fiduciary Regulation of Sexual Exploitation,’ above n 49, at 302; R Flannigan,above n 48, at 219: ‘There is no condition of liability that actors subjectively agree to thiskind of legal responsibility. It is only necessary that their arrangements, however created,involve limited access.’

52 Cf Hodgkinson v Simms (1994) 117 DLR (4th) 161, 186, [1994] 3 SCR 377 (La ForestJ).

53 Finn, above n 9, at 42. See also R Flannigan, ‘The Fiduciary Obligation,’ above n 47, at321–2; Chirnside v Fay [2004] 3 NZLR 637 (CA) 51.




the decisions and actions of those holding such offices or functions (thechance of detection of actual fiduciary breaches in individual cases islow).54

In practical terms, fiduciary loyalty is exacted through proscription of itsopposite.55 Loyalty is ensured through the operation of overlapping defaultrules—in particular the ‘no-profit’ and ‘no-conflict’ rules—that prohibitand seek to deter fiduciary disloyalty. The fiduciary is simply instructed toforgo self-interest while acting in the interests of another, at least withoutsecuring, ex ante or ex post, the genuine consent of the one to whom theduty of loyalty is owed.56 In the absence of clear proof of such consent (orother lawful authorisation)—and it is for the fiduciary positively to showthat such consent (or authorisation) was fully and freely obtained ratherthan for the obligee to show that it was not—liability is strict. There are noexcuses besides consent or authorisation for disloyal actions or tendencieswithin the scope of the fiduciary function. Indeed, the special accountabil-ity regime that responds to breaches of fiduciary obligation—strict liability,followed by the potential availability of significant remedies such asequitable compensation, constructive trust, and account of profits—servesas a legally self-conscious attempt to suppress or deter fiduciary disloyaltyby removing any conceivable incentive for the fiduciary to act inconsist-ently with his or her special duty of abstinence.57 Moreover, so strong isthe prophylactic stance here that a mere disloyal tendency in dealingsbetween a fiduciary and his or her obligee, if it falls within the scope of thefiduciary’s special function, is regarded as sufficient to justify a presump-tion of abuse and the resultant shifting of the burden onto the fiduciary toshow the propriety of the impugned dealing as a whole. In other words, thepotential of fiduciary liability is triggered by contingencies—a level ofrisk—that would indicate to a court the (realistic) possibility of abuse offiduciary office or function rather than the probability (or substantiallikelihood) of it.58 Failing clear proof of consent or authorisation, there-fore, liability might still be imposed upon a fiduciary even in the absence of

54 As Flannigan, ‘The Strict Character of Fiduciary Liability,’ above n 49, at 212,comments, ‘These considerations ground our insistence on a strict liability. Our determinationis to reduce the prospect that fiduciaries will conclude that there are opportunities to avoid orinhibit detection, or to construct plausible explanations after the fact.’ As Flannigan explains,fiduciaries are often well placed to erect plausible facades of righteousness in relation totransactions entered into with their obligees. The law’s approach to the regulation of fiduciarytasks or functions must thus be sensitive and responsive to this fact or risk.

55 Breen v Williams (1996) 186 CLR 71 (HCA) 113 (Gaudron and McHugh JJ); see alsoGummow J, ibid, at 137–8; Finn, above n 9, at 2, 28.

56 For a discussion of the defence of consent in the fiduciary context, see J Payne,‘Consent’ in P Birks and A Pretto (eds), Breach of Trust (Oxford, Hart Publishing, 2002) ch10.

57 Cf Warman International Ltd v Dwyer (1995) 182 CLR 544 (HCA) 557–8.58 The difference between ‘possibility’ regulation and ‘probability’ regulation is nicely put

by P Finn, Integrity in Government (Second Report)—Abuse of Official Trust: Conflict of

392 Rick Bigwood



direct proof of the actual influence of the conflict in the particular case, butthis is simply a concession to the strong public desire to control the verymischief and epistemological difficulties that justify imposition of fiduciaryresponsibility in the first place.

B. What Identifies the Imposition of a Fiduciary Obligation? The‘Limited-Access’ Abstraction

Courts and commentators have long struggled to articulate a clear anduniversal test for identifying a fiduciary obligation in particular instances,especially when the obligation recognised is ‘fact-based’ rather than‘status-based’. Following Robert Flannigan’s excellent work in this field,though, it can be seen that the common physical characteristic of thosearrangements in which one party will be a fiduciary relative to another is‘limited access to assets’: one party has, wholly or partly, acquired access tothe assets of another for a ‘defined or limited’ purpose, that is, for thepurpose of furthering the objectives of that other party to the exclusion ofthe first party’s (unauthorised, inconsistent) personal interests. In otherwords, access to the other’s assets is not ‘open’ as it is, for example, insimple exchange relations or encounters but rather ‘functional and con-strained’ by the relevant interests of that other party (or possibly theparties’ joint interests). The law recognises the serious mischief associatedwith limited-access arrangements—that the value of the assets can readilybe diverted or exploited for self-serving ends—and it seeks to control it byimposing, to the extent of the access, ‘fiduciary’ responsibility.59

It is to be noted that the concepts of ‘access’ and ‘assets’ in thisconnection are liberal. ‘Access to assets’ refers to access that is either direct(for example, where a person is a fiduciary because he or she is holdingtitle over another’s property in a managerial capacity) or indirect (forexample, where a person has become a disinterested adviser for anotherand so has special access to the other’s assets via his or her ‘will’ or legalcapacity for decision-making).60 The concept of ‘indirect access to assets’

Interest and Related Matters (Canberra, Australian National University, 1993) 12–13, thus:‘[Possibility regulation is] … activated characteristically by the mere existence of a particularset of stipulated conditions, irrespective of whether there are actual grounds to suspect in agiven instance that those conditions will, or are likely to, result in an abuse of office. [Incontrast, probability regulation] … ordinarily attracts the additional requirement of actualgrounds for suspicion in any given case. The practical difference between the two is that theformer produces rules which in effect presume the very phenomenon of which the latterrequires positive proof in some measure, ie the likelihood of an abuse in a given instance.’

59 See generally the writings of Flannigan, above n 49.60 Flannigan, ‘The Fiduciary Obligation,’ above n 47, at 308–309.




thus epitomises the relational undue influence situation, where the defend-ant’s transactions with the plaintiff are constrained by fiduciary responsi-bilities (disabilities) precisely because of the degree to which he or sheknowingly controls, or is capable of controlling, the will, and hence thedecision-making, of the plaintiff. The meaning of ‘assets’ is also expansive,because it comprehends the acquisition of intangibles such as contractualrights, which the defendant may acquire independently of the tangibleassets or benefits that may flow from the enforcement of contractualrights.61

C. Toward an Understanding of Relational Undue Influence Inside‘Fiduciary Law’: Defining the ‘Fiduciary Trust’ Types

Most seem to accept that the concept of ‘trust’ is somehow central to afiduciary obligation. There is lesser agreement on other descriptors thathave been employed from time to time in search of the defining character-istic or chief determinant of fiduciary status: ‘reliance’, ‘vulnerability’,62

‘reasonable (or legitimate) expectations’63 and the like. The fact is, though,that all these descriptors are unstable to the extent that each is susceptibleof overly expansive interpretations.64 For example, to say for presentpurposes that ‘P trusts D’ is unilluminating until we first understandprecisely the way in which P trusts D, and for what purposes. P must trustD in some relevant way.65 Only trust that leads to or results from D having‘limited access’ to P’s assets—access for P’s objectives rather than for D’sown—attracts a ‘fiduciary’ obligation, hence fiduciary regulation. On thisview, moreover, fiduciary trust can only ever be identified by itsconsequences—fiduciary access—and the serious risk of opportunism thatsuch access affords. Although the existence of subjective trust or confi-dence reposed in an alleged fiduciary can indicate that fiduciary accessexists in the context of the relationship or dealing under examination, it isclear that subjective trust (or expectation) is neither a necessary nor asufficient condition of fiduciary status. There is a significant prescriptive

61 Ibid, at 308, fn 118.62 See, eg, Hospital Products Ltd v United States Surgical Corporation (1984) 156 CLR

41 (HCA) 142 (Dawson J). For an account of the Supreme Court of Canada’s preoccupationwith ‘vulnerability’ as the key defining characteristic of the fiduciary relationship, see MHOgilvie, ‘Fiduciary Obligations in Canada: From Concept to Principle’ [1995] Journal ofBusiness Law 638.

63 Finn, above n 9, at 46–7; accepted by Kirby P in Breen v Williams (1994) 35 NSWLR522 (SC) 544.

64 Cf the comments of Kirby J in Pilmer v Duke Group Ltd (in liq), above n 50, at 288.65 Cf Flannigan, ‘The Fiduciary Obligation,’ above n 47, at 302: ‘Everyone is a fiduciary

who is trusted by another in the relevant way.’ Cf also Breen v Williams, above n 55, at 93(Dawson and Toohey JJ), where this point is well recognised in relation to the doctor–patientrelationship.

394 Rick Bigwood



dimension to the concept of trust in this field of law, and, regardless of theactual belief of the parties involved, limited access either exists in connec-tion with the particular task or relation, or class of tasks or relations, or itdoes not.66 Again, this is just to underscore the fact that the ‘limited-access’abstraction, rather than ‘trust’ per se, is the singular defining characteristicor prime determinant of fiduciary status.

Accepting that fiduciaries are ‘trusted’ because they have qualified(functional and constrained rather than open) access to their obligee’sassets, it next remains to distinguish between the different types of‘fiduciary trust’ (or ‘limited access’) that might exist in connection with a‘trusted’ person for the purpose of attracting fiduciary accountability. Suchdifferentiation, moreover, allows us to appreciate the sheer diversity in therange of relationships or tasks that are vulnerable to the mischief associ-ated with limited access, and hence that are subject to the imposition of afiduciary obligation. Its effect is to demonstrate how ‘fiduciary regulation’has a much wider application in our social relations or functions than somejudges and writers seem to appreciate or care to acknowledge,67 as well asto show that the ‘influential’ party in a so-called ‘relation of influence,’ forthe purposes of relational undue influence law, is invested with fiduciaryaccess and hence is subject to the usual disabilities that are engrafted uponthe transactional liberty of a self-serving fiduciary.

As Flannigan has explained, fiduciary trust may present as either of twobasic types: ‘vigilant’ trust or ‘deferential’ trust.68 Vigilant trust denotes thetype of trust that is normally reposed in intermediaries69—in those whohave an intermediary function in the employment of the trusting party’sassets, which function gives the intermediary ‘limited access’ to thoseassets. It is, accordingly, of little concern to us in connection with relationalundue influence liability. Trustees, solicitors, agents, directors, partners and(certain) employees are examples of those in whom vigilant trust istypically reposed by (respectively) beneficiaries, clients, principals, com-panies, fellow partners and employers. The fiduciary obligation is imposed

66 Again cf the comments of Kirby J in Pilmer v Duke Group Ltd (in liq), above n 50, at288.

67 Flannigan, above n 48, at 216, fn 1: ‘Most of us are accountable as fiduciaries in oneor more respects most of the time.’ And see R Flannigan, ‘Fiduciary Mechanics’ (2008) 14Canadian Labour and Employment Law Journal 25.

68 In the High Court of Australia, in Breen v Williams, above n 55, at 82, Brennan CJrecognised a distinction between fiduciary trust types that roughly parallels Flannigan’s.According to the Chief Justice, fiduciary duties arise from two distinct but frequentlyoverlapping sources: (a) agency, and (b) relationships of ascendancy or influence by one partyover another or dependence or trust on the part of that other.

69 This label is chosen by Flannigan because those who repose ‘vigilant trust’ in otherstend to remain wary or ‘vigilant’ that their trust may at any time be misplaced. Employers inrespect of their servants or agents, for example, are entitled to expect that the latter will servethe former faithfully, although unreserved or unsceptical faith is rarely conceded as it is in the‘deferential trust’ cases. See Flannigan, ‘The Fiduciary Obligation,’ above n 47, at 287.




here on the basis of the ‘intermediary [or ‘agency’] cost mischief’70—the‘cost’ to the trusting party of the mischief that arises when beneficialownership is separated from management and control.71 The fiduciaryobligation is thus imposed in an attempt to avoid those costs: that is, inorder to deter opportunistic conduct.

The basis for recognising (imposing) a fiduciary obligation in cases thatinvolve deferential trust is rather different from the intermediary costjustification that explains such an obligation in the vigilant trust cases.According to Flannigan, deferential trust is ‘deferential’ in the sense that‘the trusting person will defer to the judgment of the trusted person’.72

Sometimes the deference involved is total, while at other times it is merelypartial or situational.73 In some cases such as Johnson v Buttress,74 to bediscussed below, it is attended by elements of necessity, dependence orsubmission. But in other cases no such obvious vulnerability exists, yet stillthe ascendant party knows that his or her impartial judgment and guidanceis being relied on in the circumstances.75 In every case it is necessary toexamine the relationship between the parties involved, and in particularthe relative positions and precise roles within that relationship. To this end,one party’s role or position must be such as to ‘implicate [him] in theother’s affairs or so align him with the protection or advancement of thatother’s interests that foundation exists for the “fiduciary expectation.”’76

Such a role may exist, or be taken to have arisen, in a variety of ways andin a variety of settings,77 but typically it is found ‘either because [the

70 ‘Intermediary’ or ‘agency’ costs are defined in different ways by different commenta-tors. See Flannigan, ‘The Economics of Fiduciary Accountability,’ above n 49, at 397, fn 12.For Flannigan, ‘agency costs are the costs of opportunism and the costs of controllingopportunism’: ibid.

71 See Flannigan, ibid, at 394–401. See also Duggan, above n 47, at 429–31.72 Flannigan, ‘The Fiduciary Obligation,’ above n 47, at 286. Cf also the concept of

‘reverence’ in Continental law: J du Plessis and R Zimmermann, ‘The Relevance of Reverence:Undue Influence Civilian Style’ (2003) 10 Maastricht Journal of European and ComparativeLaw 345.

73 Ibid.74 Above n 15.75 Flannigan cites as examples Huguenin v Baseley, above n 44, at 293–4 (Lord Eldon

LC) and Lloyds Bank Ltd v Bundy, above n 44, at 341 (Sir Eric Sachs): see Flannigan, ‘TheFiduciary Obligation,’ above n 47, at 287, fn 18. Commentators have fairly doubted whetherLloyds Bank Ltd v Bundy can be supported on a fiduciary (qua relational undue influence)basis. See, eg, P Finn, ‘Contract and the Fiduciary Principle’ (1989) 12 University of NewSouth Wales Law Journal 76, 96, arguing that the case is supportable on an unconscionabledealing basis only. Eisenberg also regards this as an unconscionability (qua ‘transactionalincapacity’) case: ‘the real vice of the transaction was that the bank had led a relativelyunsophisticated person into a transaction with a severe though not readily apparent potentialfor unfairness, without pointing out the need for expert advice’: see MA Eisenberg, ‘TheBargain Principle and Its Limits’ (1982) 95 Harvard Law Review 741, 767.

76 Finn, above n 9, at 47.77 For example, it may exist or arise by nature, independent of human will, such as in the

parental or quasi-parental function, or it may have been brought about by the voluntary

396 Rick Bigwood



defendant] is or has become an adviser of [the plaintiff] or because he hasbeen entrusted with the management of [the plaintiff’s] affairs or everydayneeds or for some other reason,’ and so ‘is in a position to influence [theplaintiff] into effecting the transaction of which complaint is later made’.78

The effect of the parties’ relation or circumstances is always ‘deference’ inthe sense that what follows is ‘relaxation of [the plaintiff]’s self-interestedvigilance or independent judgment in favour of [the defendant’s] protectionor judgment because the circumstances of the relationship justify the beliefthat [the defendant] is acting or will act in … [the plaintiff]’s interests’.79

Deferential trust, then, is trust that gives the trusted party special abilityto influence the trusting party’s decisions: it gives the former indirect accessto the latter’s assets, which is access through the latter’s ‘will’ or capacity tomake decisions. The former effectively has control over the latter’sdecision-making in connection with matters that fall within the scope ofthe parties’ deferentially trusting relationship or encounter. Deferentialtrust differs from vigilant trust only in the manner in which it arises andsubsequently operates as between the parties involved. Vigilant trustgenerally arises because the plaintiff has made a decision that he or sheexpects the defendant as an intermediary to implement (an ‘implementa-tional’ trust, says Flannigan80), whereas deferential trust arises because thedefendant is, or can be taken to be, serving as a disinterested adviser orconfidant to the plaintiff, or because the defendant knowingly occupies aposition of authority, respect or expertise relative to the plaintiff.81 Vigilanttrust thus operates in association with the implementation of decisions bythe defendant as the trusted party, but it does not necessarily comprehendthe defendant having an opportunity to participate in the original decision-making (by or on behalf of the plaintiff) that gave rise to the implementa-tional trust itself. Deferential trust, in contrast, envisages input by thedefendant at the decision-making stage, that is, at the moment when thedefendant enters into a beneficial transaction with the plaintiff.82 Althoughat that moment it may appear to an objective observer that the plaintiff ismaking his or her own decisions, in reality it is, or might be, thedefendant’s will that is operative.

Note that although vigilant trust and deferential trust differ in theseimportant respects, of which more below, they are otherwise identical in

action of the parties themselves, such as in the professional relations of solicitor and client, orit may have originated by circumstances and conduct generally, such as in Johnson v Buttress(discussed below at text accompanying nn 84–95).

78 Goldsworthy v Brickell, above n 44, at 401 (Nourse LJ).79 Finn, above n 9, at 48.80 Flannigan, ‘The Fiduciary Obligation,’ above n 47, at 293.81 Ibid, at 292–3.82 Ibid, at 293. Flannigan calls this ‘decisional’ trust, in contrast to ‘implementational’

trust.




their essential ‘trust’ quality for the purpose of attracting ‘fiduciary’regulation in respect of certain dealings between the plaintiff and thedefendant, or between the defendant and a third party. This is becauserecognition of a fiduciary obligation in both cases is an attempt to avoidthe self-same mischief of fiduciary opportunism in limited-access arrange-ments. The risk in both situations is that the defendant’s commitment tothe defined and limited purpose of the relationship with the plaintiff will becompromised by the prospect of personal gain to the defendant (or to athird person at the defendant’s direction). Although the source of theopportunity for abuse might differ as between the two forms of fiduciarytrust—the separation of asset ownership from management and control inthe case of vigilant trust, and the special capacity to influence the trustingparty’s decision-making in the case of deferential trust—the opportunitiesfor such abuse are nevertheless identical:

Both types of trust in fact result in the trusted party acquiring ‘access’ to theemployment of assets. In the case of deferential trust, however, the access isindirect because it occurs through ‘influence’ exerted by the trusted party. But ineither case, and to the same extent, the ‘access’ to assets may be turned tomischievous ends.83

D. A Consummate Judicial Encapsulation of Relational Undue Influenceas ‘Fiduciary’ Regulation: Dixon J in Johnson v Buttress

Johnson v Buttress was decided by the High Court of Australia in 1936.84

The respondent’s father, Buttress, had made to the appellant, Mrs Johnson,an inter vivos gift of a block of land on which stood a cottage. AfterButtress’s death, the respondent challenged the gift on the ground of undueinfluence. At the time of the gift Buttress was of advanced years andrecently widowed. He was illiterate and peculiarly dependent upon others.Mrs Johnson, who was a distant relative of Buttress’s wife and a long-timefriend of the couple, had remained close to Buttress after the wife’s deathand generally took care of him. The majority of the court found thatButtress, by reason of his age, illiteracy, temperament, inferior mentalfaculties and particular mode of life, had come to depend on Mrs Johnstoncompletely for her guidance, support, supervision and kindness.85 Dixon J,in particular, held that the circumstances showed that Mrs Johnson had, in

83 Ibid, at 309.84 Above n 15.85 Latham CJ, Dixon J and McTiernan J delivered separate judgments to the same effect.

Evatt J agreed with Dixon J’s judgment. Starke J, in contrast, had difficulty in accepting thatthe facts disclosed a peculiar relation of trust and confidence between the donor and MrsJohnson so as to bring him within the ‘protected class’ in respect of which there is apresumption of undue influence. Nonetheless, Starke J thought that the facts afforded

398 Rick Bigwood



the time leading up to the impugned gift, in fact stood in an ‘antecedentrelation of influence’ to Buttress. Although there was in his view no directproof that the transaction was procured by the improper exercise of anactual ascendancy or domination gained by Mrs Johnson over Buttress, thefact that a substantial gift had been made by Buttress to Mrs Johnsonwithin the scope of the parties’ special antecedent relation of influenceimposed on the latter the ‘burden of justifying the transfer by showing thatit was the result of the free exercise of [Buttress’s] independent will’—aburden that, in Dixon J’s view, Mrs Johnson had ‘quite failed’ to dis-charge.86

For Dixon J, the foundation of equity’s jurisdiction to rescind a transferof property for undue influence lay in ‘the prevention of an unconscien-tious use of any special capacity or opportunity that may exist or arise ofaffecting the alienor’s will or freedom of judgment in reference to such amatter’.87 As was orthodoxy at that time, and indeed is still now afterEtridge, Dixon J divided undue influence into two distinct categoriesaccording to the source of the power to exercise control over (that is, to‘influence’) the alienor’s will or freedom of judgment. The first categoryencompassed those cases where the source of the power lay in ‘noantecedent relation but in a particular situation, or in the deliberatecontrivance of the dominant party,’88 although such a power may coinci-dentally arise in the context of such an antecedent relation. Cases fallingwithin this category would today be recognised as Class 1, ‘actual’ undueinfluence claims, in which the party who could establish such a power oropportunity and who wished to upset the transaction would bear the onusof adducing further evidence to show that the transaction was the outcomeof the influence having actually been exerted over the mind or will of thesubordinate party.

The second category of undue influence identified by Dixon J encom-passed those cases where the source of the power to exercise control overthe alienor’s will or freedom of judgment lay distinctively in the situationwhere ‘the parties … antecedently stand in a relation that gives to one anauthority or influence over the other from the abuse of which it is properthat he should be protected’.89 Where the parties stood in such a relation,the influential party who received ‘property of substantial value’, forexample by way of gift, from the other party could not retain the beneficialtitle to the subject matter of the gift ‘unless he satisfie[d] the court that he

sufficient evidence to allow a court justly to infer that the transfer was involuntary as havingresulted from ‘unfair and undue pressure’ on the part of Mrs Johnson.

86 Johnson v Buttress, above n 15, at 138. Cf McTiernan J, ibid, at 143.87 Ibid, at 134.88 Ibid.89 Ibid.




took no advantage of the donor, but that the gift was the independent andwell-understood act of a man in a position to exercise a free judgmentbased on information as full as that of the donee’.90 Such a burden fellupon the one who received substantial benefits within the context ofcertain well-known relationships that ‘by their very nature import influ-ence’. (These would be known today as ‘Class 2(A)’ relationships ofinfluence.) By way of example Dixon J mentioned the familiar relations ofsolicitor–client, parent–child, physician–patient, guardian–ward, andfiancé–fiancée, but he was also quick to indicate that the burden ofjustifying the transaction was not confined to fixed categories. Rather, itrested upon a principle, and the fiduciary principle no less:

It applies whenever one party occupies or assumes towards another a positionnaturally involving an ascendancy or influence over that other, or a dependenceor trust on his part. One occupying such a position falls under a duty in whichfiduciary characteristics may be seen. It is his duty to use his position of influencein the interest of no one but the man who is governed by his judgment, gives himhis dependence and entrusts him with his welfare.91

When such a party took a substantial gift of property from the dependentor trusting party, Dixon J explained, or entered into a transaction that‘wears the appearance of a business dealing’, a presumption against thevalidity of the disposition arises in the latter party’s favour, so that itbecomes ‘incumbent upon [the transferee] to show that it cannot beascribed to the inequality between them which must arise from his specialposition’.92 Such a presumption, in Dixon J’s view, rested upon a ‘firmfoundation’ of policy, and upon other considerations that combined tojustify strict fiduciary regulation, namely:93

— that the transferee may be taken to possess a peculiar knowledge notonly of the disposition itself but also of the circumstances that shouldaffect its validity;

— that the transferee chose to accept a benefit that may well proceedfrom an abuse of the authority conceded to him or her, or the authorityreposed in him or her; and

— that the relations between the transferor and the transferee are so closeas to render it difficult to disentangle the inducements that led to thetransaction.

Justice Dixon also emphasised that this rule must not be narrowed: ‘therisk must not be run of fettering the exercise of the jurisdiction by an

90 Ibid.91 Ibid, at 134–5.92 Ibid, at 135.93 Ibid.

400 Rick Bigwood



enumeration of persons against whom it should be exercised; the reliefstands upon the general principle applying to all the variety of relations inwhich dominion may be exercised by one person over another.’94

To my mind, Dixon J’s judgment in Johnson v Buttress is the mostpellucid judicial elaboration of the essential ‘fiduciary’ nature of relationalundue influence that exists in the law reports of the major BritishCommonwealth legal systems.95 It signifies that, in purely functional terms,the equitable jurisdiction to rescind transactions on the basis of Class 2undue influence is indistinguishable from all other categories (such asbreach of confidence and wider fiduciary obligation) that are united by thesingular and generic goal of controlling opportunism in limited-accessarrangements. Separate doctrinal administration by the courts should notbe taken to imply that different conceptual or substantive concerns arenecessarily involved in each situation. The interests protected, and thebroad mischief sought to be avoided, are identical as between Class 2undue influence and wider fiduciary law. Both fall squarely within theambit of the same policy goal of maintaining the integrity of ‘trusting’relations or institutions; both are ‘fiduciary’ for the same reason, that is,because one party has acquired ‘limited access’ to another’s assets, whichraises a singular opportunism concern, and which in turn justifies theimposition of ‘fiduciary’ accountability to the extent of the limited access.The mechanistic difference with relational undue influence is the means by

94 Ibid, at 136, clearly echoing the words of Chelmsford LC in Tate v Williamson, aboven 44, at 61.

95 Sir Eric Sachs’ judgment in Lloyds Bank Ltd v Bundy, above n 44, and Dunn and SladeLJJ’s judgments in Morgan, above n 30, in the English Court of Appeal are also instructive.Although the law of undue influence appears to differ from state to state in the United States,attention should be paid to Martinelli v Bridgeport Roman Catholic Diocesan Corp 196 F 3d409 (2d Cir 1999), in which the court, applying Connecticut law, observed that theconventional view is that the normal rule (that a person alleging a wrong must prove it) ‘issomewhat relaxed in cases where a fiduciary relation exists between the parties to atransaction or contract, and where one has a dominant or controlling force or influence overthe other. In such cases, if the superior party obtains a possible benefit, equity raises apresumption against the validity of the transaction or contract, and casts upon such party theburden of proving fairness, honesty, and integrity in the transaction or contract … Therefore,it is only when the confidential relation is shown together with suspicious circumstances, orwhere there is a transaction, contract, or transfer between persons in a confidential orfiduciary relationship, and where the dominant party if the beneficiary of the transaction,contract, or transfer, that the burden shifts to the fiduciary to prove fair dealing. A fiduciaryseeking to profit by a transaction with the one who confided in him had the burden ofshowing that he has not taken advantage of his influence or knowledge and that thearrangement is fair and conscientious’: ibid, at 421, fn 6. The court also indicated that‘Shifting the burden of proof protects fiduciary relationships by helping to ensure that thefiduciary acts consistently with the responsibilities such relationships entail’: ibid, at 421. Itdid note that ‘To be sure, where the fiduciary has not received some kind of benefit thatwould engender suspicion and there is no other evidence of wrongdoing, the burden of proofremains on the plaintiff’, ibid. As to the proof required, the court held that ‘as in otherinstances in which the burden shifts to the fiduciary to show fair dealing, such proof must beby “clear and convincing evidence”’, ibid, at 423.




which the fiduciary’s office or function is, or can be, misused. In asuccessful relational undue influence claim, it is the ascendant party’s‘influence’ over the deferentially trusting party, rather than a directdiversion of the value of assets, that is the mechanism of the opportunismthat lies at the foundation of the latter party’s claim for exculpation fromthe impugned transaction.

E. Respecting Administrative Differences: On Aligning ‘Obligation’ with‘Factual Structure’

The last point in the preceding section is a vital one, for although fiduciaryobligation and relational undue influence serve in common the elementalfunction of controlling opportunism in limited-access arrangements, it isnecessary nonetheless to observe important factual and practical distinc-tions that exist in connection with the administration of the respectivebodies of doctrine. These distinctions, however, merely symbolise contex-tual regulation of an otherwise conceptually unified subject matter. Theyare directed not at the nature of the obligation, regulation and liabilitiesinvolved, as those are identical across the various fiduciary categories, butrather at the possible working relationship between the various kinds ofcorrective rules and responses that are united under the rubric of a singularfiduciary regulatory framework. It is, in other words, imperative todistinguish between different kinds or modes of fiduciary breach, and tomatch the content or extent of a fiduciary’s obligation with the actualstructural characteristics of the particular physical arrangement underexamination.96 This is because the content or extent of a fiduciary’sresponsibilities, and the law’s corrective response to any actual or possiblebreach of those responsibilities, are affected by, and hence must vary with,the particular powers and opportunities enjoyed by the fiduciary, or type offiduciary, in question. Needless to say, being a ‘fiduciary’ does not subjectone to every incident of the express trust, for it is unnecessary to imposeany higher obligation on a fiduciary than is needed to maintain theintegrity of the particular limited-access relation, be it status-based orfact-based in nature.

The process of matching content with factual structure holds importantinsights for understanding Class 2 ‘relational’ undue influence as it existedbefore 1985 in the United Kingdom. In particular, it assists in locating thatform of undue influence within the wider structure of fiduciary regulation,

96 A process that Flannigan calls ‘matching content with structure’: Flannigan, ‘TheFiduciary Obligation,’ above n 47, at 320. The idea essentially refers to the assignment ofaccountability over the full range of the limited access in a particular arrangement, withoutgoing beyond that access. Cf also Attorney-General v Blake (1997) [1998] Ch 439 (CA)454–5 (Lord Woolf MR).

402 Rick Bigwood



while defensibly maintaining its discrete doctrinal administration separatefrom general fiduciary law. Although both vigilant trust and deferentialtrust are forms of ‘fiduciary’ trust by virtue of the fact that each leads to‘fiduciary access’ and hence to a common toxic mischief that the law seeksto avoid, this is not to say that the legal scope of protection afforded to therespective trust types must be identical for all intents and purposes.Consider vigilant trust. Where vigilant trust has been reposed in afiduciary, liability is unconditionally strict. This is because vigilant trust iscapable of being breached in many and varied ways, all of which aregenerally unappreciated by the trusting party or difficult to detect. The lawaccordingly holds the vigilantly trusted party strictly liable for all actual orpotential conflictual acts or states of affairs that fall within the scope of thetrusting relation or function, no matter how minor or innocent such acts orstates of affairs may be. Moreover, when the victim of a breach of vigilanttrust seeks transaction avoidance (as opposed to, say, a personal account-ing or compensatory remedy),97 causation between the breach and theimpugned transaction is irrelevant from the standpoint of the victim’sentitlement to rescind. That is because unauthorised entry into a trans-action that is inconsistent with the fiduciary’s duty of loyalty is itself abreach of duty ‘patent at the creation of the very thing which is to be setaside’.98

Turning to deferential trust, or limited-access relations of influence, adifferent legal approach is evident. The focus is now on the specialinfluential capacity enjoyed by the ascendant party in virtue of thesubordinate party’s known deference toward the former in relation tomatters falling within the area of his or her authority, expertise orresponsibility. In contrast to the vigilant trust situation, therefore, there isonly one possible form of objectionable conduct that can occur: theeffective and conflictual exercise by the fiduciary of that influence over thesubordinate party for the fiduciary’s own ends, which, at least pre-Morgan,is what defined all pure acts of relational undue influence. When such a useof influence causes the impugned transaction, the consent brought to thattransaction by the subordinate party cannot be treated as a expression ofthat party’s ‘free and informed’ will, so as to render him or her fully legallyresponsible for that consent, as his or her will was compromised by theunauthorised conflictual use of the ascendant party’s special influence.

97 It has been argued that causation should also not be required when an accountingremedy is sought after breach of the no-profit rule: see V Vann, ‘Causation and Breach ofFiduciary Duty’ [2006] Singapore Journal of Legal Studies 86. Cf Regal (Hastings) Ltd vGulliver [1967] 2 AC 134 (HL) 144; Gray v New Augarita Porcupine Mines Ltd [1952] 3DLR 1 (PC) 15. Views have been expressed to the opposite effect: see C Mitchell, ‘Causation,Remoteness, and Fiduciary Gains’ (2006) 17 Kings College Law Journal 325; Chirnside vFay, above n 53 (concerning a joint venture).

98 Maguire v Makaronis (1997) 188 CLR 449 (HCA) 467.




Causation now becomes relevant because, unlike in the vigilant trustsituation, limited-access influence that results from deferential trust canonly be effectuated instrumentally through the deferentially trusting party,who must thus be involved at the critical decision-making (but notnecessarily the implementational) stage—that is, at the time of transacting.It may be that, despite the defendant’s apparent or possible use of specialinfluence over the plaintiff to effectuate the impugned transaction, thelatter actually made the transactional decision, independently of theformer’s influence. In other words, the plaintiff may have in fact actedindependently, either because the defendant did not actually exerciseinfluence over the plaintiff or, if he or she did attempt to do so, thisinfluence was ineffectual in bringing about the impugned transaction.Fiduciary liability may also be less strict in the deferential trust situationthan in the vigilant trust cases, as even if the defendant had in fact enjoyedfiduciary influence over the plaintiff that was effectual in producing theimpugned transaction, some formulations of the relational undue influencedoctrine suggest that the defendant may still be absolved of liability (andthus retain any value received from the plaintiff) by showing that no unfairprocedural or strategic advantage had been taken of that influence.99 Thatmight occur, for instance, because the defendant was, at the time of thetransaction, justifiably ignorant of the plaintiff’s deferential trust towardthe defendant, or was entitled reasonably to believe that the plaintiff hadbeen independently and competently advised on both the conflict and thegeneral wisdom of what he or she was about to do. In other words, anyconflict in the relational undue influence context is managed through a‘fair-dealing’ rule, as an application of the overarching no-conflict rule,rather than a pure ‘self-dealing’ rule.100

Yet for all that, care must be taken not to overstate the contrast, sincethe defence to an alleged breach of trust, whether vigilant or deferential,

99 I should emphasise that ‘procedural fairness’ must be shown, and that merely showing‘substantive fairness’ does not alone suffice to avoid liability in this context (althoughsubstantive fairness may bolster a defence of procedural fairness). To be sure, giving too muchweight to the existence of substantive fairness in the fiduciary context raises the usualdetection and evidentiary concerns, in that it encourages the possibility of cosmeticallymanipulated transactions.

100 Note that there is a parallel type of analysis in relation to the ‘self-dealing’ and‘fair-dealing’ rules where trustees purchase trust property or a beneficiary’s beneficial interest.Despite Megarry V-C’s suggestion to the contrary in Tito v Waddell (No 2) [1977] Ch 106,225, 241, both of those rules should now be seen as contextual regulation of the genericno-conflict rule: see, eg, E Simpson, ‘Conflicts’ in Birks, above n 56, at ch 3, especially at 82ff;M Conaglen, ‘A Re-Appraisal of the Fiduciary Self-Dealing and Fair-Dealing Rules’ [2006]CLJ 366. Note also that although relational undue influence does not involve ‘self-dealing’ inits purest sense (that is, where the selling/purchasing fiduciary controls the subject matter onboth sides of the transaction), self-dealing nevertheless does occur in the sense and to theextent that the fiduciary’s influence operates to control both sides of the transaction. CfShepherd, above n 18, at 156.

404 Rick Bigwood



remains identical in both situations, namely, that the plaintiff ‘fullyconsented’ to the benefit received, which consent must include the plaintiffhaving received disclosure of the conflict that would otherwise exist. Howthe different types of trust operate, however, determines the different stepsthat the defendant must take toward securing the plaintiff’s consent in thatregard. Only undue influence requires the act of emancipation that, in apractical sense, operates to eliminate the conflictual state of affairs, as wellas securing the plaintiff’s personal transactional consent to the impugnedbenefit. This should not, however, be understood as implying that deferen-tial trust is somehow less worthy of protection than vigilant trust, or thatthe legal scope of protection is somehow narrower. The differencesbetween the respective forms of fiduciary regulation and specific legalresponses are simply consequences of the varying physical phenomena thatattend the parties’ relationship or arrangement in each situation: the natureof the different types of limited access (or fiduciary trust) involved, themanner in which each type of access or trust tends to operate, the specifictypes of opportunity that each tends to present to the fiduciary fordiverting value to himself or herself (or to third parties), and the specificand varying forms of wrongdoing (opportunism) that might relevantlyflow from such opportunities.

III. REFLECTIONS ON THE DISINTEGRATION OF RELATIONAL UNDUEINFLUENCE AS A FIDUCIARY RULE—FROM MORGAN TO ETRIDGE

IN THE HOUSE OF LORDS

A. National Westminster Bank plc v Morgan

As mentioned earlier, the essential fiduciary character of relational undueinfluence in the United Kingdom was suppressed or diminished in 1985with the unanimous decision of the House of Lords in Morgan. There theLaw Lords expressly rejected the so-called ‘public policy’ approach torelational undue influence. Lord Scarman, with whom the other Law Lordsagreed, held that the basis for relief in this area was ‘not a vague “publicpolicy” but specifically the victimisation of one party by the other’.101 Onthis approach, the party seeking exculpation from the transaction must, inaddition to proving a special relationship of influence inter partes, showthat the resultant bargain was ‘manifestly disadvantageous’ to him or her,

101 Above n 30, at 705, citing Lindley LJ in Allcard v Skinner, above n 7, at 182–3.Interestingly, in Etridge the House of Lords did not comment directly on the compatibility (orotherwise) of the two approaches. However, the language employed in the various judgmentsis entirely consistent with that of the ‘victimisation’ approach and the ‘public policy’ approachis not discussed at any point.




as judged by the (obviously) substantively unfair nature of the impugnedtransaction itself. No presumption of undue influence could therefore arisein respect of a transaction that provided ‘reasonably equal benefits forboth parties’.102 His Lordship also preferred to ‘avoid the term “confiden-tiality” as a description of the relationship which has to be proved’ in thepresumptive (Class 2) cases.103

Lord Scarman’s judgment, however, is problematic in many respects.Although some of his errors were identified and corrected in later House ofLords decisions—of which more shortly—the effects of the dismissal of the‘public policy’ (fiduciary) approach to relational undue influence in Mor-gan are still evident in the subsequent enunciation of the ‘first principles’ ofundue influence in Etridge. This is despite the fact that, amid those twodecisions, Lord Browne-Wilkinson, in Pitt,104 flagged the possible need toconsider ‘the exact limits of Morgan … in the future’. In particular hisLordship said:

The difficulty is to establish the relationship between the law [of undueinfluence] laid down in Morgan and the long established principle laid down inthe abuse of confidence cases viz. the law requires those in a fiduciary positionwho enter into transactions with those to whom they owe fiduciary duties toestablish affirmatively that the transaction is a fair one … [This] principle isfounded on considerations of general public policy, viz. that in order to protectthose to whom fiduciaries owe duties as a class from exploitation by fiduciariesas a class, the law imposes a heavy duty on fiduciaries to show the righteousnessof the transactions they enter into with those to whom they owe such duties. …Unfortunately, the attention of this House in Morgan was not drawn to theabuse of confidence cases and therefore the interaction between the twoprinciples (if indeed they are two separate principles) remains obscure105

As we shall see, the House of Lords subsequently did not, in Etridge,confront this ‘difficulty’ or attempt directly to resolve the continuing‘obscurity’ between the two principles.

Turning now to Lord Scarman’s reasoning in Morgan, and in particularhis statement that the principle justifying interference with a transactionfor undue influence is ‘victimisation’ rather than ‘vague “public policy,”’ itmust first be noted that his Lordship relied for support on a series ofpassages from Lindley LJ’s judgment in Allcard v Skinner.106 In onepassage Lindley LJ opined that the undue influence doctrine is founded onthe principle that ‘it is right and expedient to save [people] from beingvictimised by other people’, rather than on the ground that they should

102 Morgan, above n 30, at 704.103 Ibid, at 709.104 Above n 7, at 209.105 Ibid, at 209 (emphasis in original).106 Above n 7.

406 Rick Bigwood



simply be saved ‘from the consequences of their own folly’.107 Pausing herefor a moment, it is obvious that this passage does not alone necessitaterejection of the ‘public policy’ approach to relational undue influence, forthat approach is itself concerned entirely with regulating against the risk ofvictimisation by influential fiduciaries in circumstances where proof ofactual victimisation by way of relational undue influence is typicallydifficult if not impossible. Although it might produce a quite artificialfinding of victimisation in a particular case (when in fact the influentialparty may have been factually innocent), in no way is the public policyapproach motivated by any desire to save parties from bad bargains orregretted gifts. On the contrary, as acknowledged in the above passagefrom Lord Browne-Wilkinson’s judgment in Pitt, the public policyapproach in fiduciary contexts exists precisely ‘in order to protect those towhom fiduciaries owe duties as a class from exploitation by fiduciaries as aclass’. The ‘public policy’ and ‘victimisation’ approaches are not, in fact,mutually exclusive of each other.108 Public policy merely justifies and isexpressed through the distinctive process of reasoning from the duty (thelimited-access relation of influence) to the breach (abuse of that relation orvictimisation in fact), that is, through a presumption of undue influencebased on the mere existence of a transaction inter se, together with the factthat ‘the character of the relation itself is never enough to explain thetransaction and to account for it without suspicion of confidenceabused’.109

The objection to Lord Scarman’s use of Lindley LJ’s distinction betweenvictimisation and mere regretted transactions aside, once making the moveto factual victimisation for additional reasons,110 his Lordship nonethelessproceeded to secure his error in rejecting the public policy approach torelational undue influence by making it clear in his judgment that heviewed the wrong of ‘victimisation’ as residing, at least in part, in thesubstance of the impugned transaction itself. I say ‘at least in part’ herebecause Lord Scarman’s judgment is actually unstable on this point. Itvacillates unsteadily, and seemingly without awareness, between, on theone hand, viewing substantive unfairness (or ‘manifest disadvantage’)merely as part of the factual matrix that would leave ‘room for the court topresume that [the impugned transaction] resulted from the exercise of

107 Ibid, at 182–3, cited in Morgan, above n 30, at 705.108 Cf Sheridan, above n 12, at 87: ‘fraud and things of confidence do not constitute a

mutually exclusive dichotomy’.109 Yerkey v Jones (1939) 63 CLR 649 (HCA) 675 (Dixon J).110 That is to say, in addition to relying on Lindley LJ’s remarks in Allcard v Skinner, Lord

Scarman relied on two Privy Council decisions—Bank of Montreal v Stuart [1911] AC 120(HL) 137 and Poosathurai v Kannappa Chettiar (1919) LR 47 IA 1 (PC) 3—which heconsidered supported his particular victimisation approach: see Morgan, above n 30, at706–707.




undue influence’111—that is, it would allow the court to infer victimisationin fact, or generate a ‘presumption’ of undue influence—and, on the otherhand, regarding it as an ingredient of the wrong of ‘undue influence’ itself.In Lord Scarman’s words, there is a:

need to show that the transaction is wrongful in the sense explained by LindleyLJ [in Allcard v Skinner—namely, that ‘the gift is so large as not to be reasonablyaccounted for on the ground of friendship, relationship, charity, or otherordinary motives on which ordinary men act’112—] before the court will set asidea transaction whether relying on evidence or the presumption of the exercise ofundue influence.113

Although these words were later to steer lower courts into error, inparticular by causing them to require proof of manifest disadvantage inrelation to Class 1 non-relational undue influence claims,114 the House ofLords promptly took corrective action in Pitt, pointing out that LordScarman’s speech ‘was primarily concerned to establish that disadvantagehad to be shown, not as a constituent element of the cause of action forundue influence, but in order to raise a presumption of undue influence[within] Class 2’.115 Lord Scott of Foscote also noted in Etridge theobvious circularity in Lord Scarman’s reasoning that no ‘presumption ofundue influence [could] arise from the evidence of the relationship of theparties without also evidence that the transaction itself was wrongful inthat it constituted an advantage taken of the person subjected to theinfluence which, failing proof to the contrary, was explicable only on thebasis that undue influence had been exercised to procure it’.116 How coulda transaction be adjudged ‘wrongful’ before it was proven to have beenprocured by undue influence?

Now, regardless of what it is that triggers the presumption of undueinfluence in the Class 2 cases,117 it is clear that Lord Scarman did notformulate his manifest disadvantage test by reference to the fiduciaryrationale that had, until his own rejection of the ‘public policy’ approach,governed the operation of the presumption according to the substantivepolicy that justified the creation and application of the presumption in thefirst place. Lord Scarman did not, after all, consider the influential relationbetween the parties to a Class 2 undue influence claim to be ‘fiduciary’ innature at all. Commenting adversely on the Court of Appeal’s factual

111 Morgan, above n 30, at 707.112 Above, n 7, at 185.113 Morgan, above n 30, at 706.114 See, eg, Bank of Credit and Commerce International SA v Aboody (1988) [1990] 1 QB

923 (CA) 967; Contractors Bonding Ltd v Snee [1992] 2 NZLR 157 (CA).115 Above n 7, at 208.116 Above n 6, at [155].117 I have discussed this elsewhere in Bigwood, above n 40, at 423–56.

408 Rick Bigwood



classification of the relationship between the bank and Mrs Morgan asbeing one of ‘confidence in which [Mrs Morgan] was relying on the bankmanager’s advice’, and that ‘the confidentiality of the relationship was suchas to impose upon [the manager] a “fiduciary duty of care,”’118 LordScarman said:

I believe that the Lords Justices were led into a misinterpretation of the facts bytheir use, as is all too frequent in this branch of the law, of words and phrasessuch as ‘confidence,’ ‘confidentiality,’ ‘fiduciary duty.’ There are plenty ofconfidential relationships which do not give rise to the presumption of undueinfluence (a notable example is that of husband and wife …); and there areplenty of non-confidential relationships in which one person relies upon theadvice of another, eg many contracts for the sale of goods.119

He then concluded his judgment by declaring a preference for avoidance ofthe term ‘confidentiality’ as a description of the relationship that must beproved in the presumptive cases;120 the search, rather, was for a ‘dominat-ing influence of one over another,’ which search required a ‘meticulousexamination of the facts’.121

The problem with Lord Scarman’s criticism of the Court of Appeal’s useof ‘fiduciary’ terminology in connection with the influential relation thatmust be shown in the presumptive cases is that it simply begs the questionas to where the real points of similarity and difference lie betweenrelational undue influence and wider fiduciary law. His reference to‘reliance upon the advice of another’, for example, is simply unhelpfulunless we first agree that we are only talking about reliance in the relevantsense, as importing something that leads to ‘limited access to assets’.122

‘Reliance’ is open to expansive meanings and is thus too unstable as adescriptor for identifying a fiduciary obligation. And even assuming thatthe relation of husband and wife can be described as a ‘confidential’ one(again, in the relevant sense),123 the reason why the presumption of undueinfluence does not apply to transactions between spouses relates notnecessarily to the nature of the relationship involved but rather to the lackof natural suspicion surrounding generous dispositions inter se. In other

118 Morgan, above n 30, at 702.119 Ibid, at 703.120 Ibid, at 709.121 Ibid.122 Taking Lord Scarman’s ‘sale of goods’ example, for instance, although it is clear that I

might rely on a seller’s advice (eg that the goods I wish to acquire will suit my purposes orperform in a certain way), the seller’s access to my assets in exchange for those goods is ‘open’or ‘unlimited’. In other words, I do not, or am not entitled to, rely on the seller to suspendself-interest in connection with the relationship or encounter in question, as its access to mymoney in return for the goods is for its own account and not mine.

123 That is, as importing ‘some quality beyond that inherent in the confidence that canwell exist between trustworthy persons who in business affairs deal with each other at arm’slength’: Lloyds Bank Ltd v Bundy, above n 44, at 341 (Sir Eric Sachs).




words, partiality is expected between spouses (and similarly intimateparties); there is ordinarily nothing that prima facie ‘calls for explanation’so as to trigger the presumption of undue influence.124

Indeed, the sort of question-begging evident in Lord Scarman’s judgmentseems to contaminate all denials of the essential ‘fiduciary’ nature ofrelational undue influence, whether they are absolute in nature125 orqualified in some way (typically by the rider that the two bodies ofdoctrine ‘overlap but do not coincide’).126 For myself, I find all attempts atdifferentiation (so far) to be unconvincing, at least when they are directedat the conceptual level. The points of distinction are either asserted orpurely formal in nature. One cannot ignore the ubiquity of fiduciaryobligation in our society, as Fletcher Moulton LJ’s famous ‘errand boy’example in Re Coomber127 made clear,128 and in particular the fact that

124 Bank of Montreal v Stuart, above n 110; Yerkey v Jones, above n 109; Howes v Bishopand Wife [1909] 2 KB 390 (CA) 396 (Lord Alverstone), 400–403 (Farwell J); Etridge, aboven 6, at [19] (Lord Nicholls). Additional phenomena may attend the husband and wiferelationship so as to make it a Class 2(B) situation: Etridge, above n 6, at [130] (LordHobhouse), [283] (Lord Scott, on Barclays Bank plc v Coleman [2001] QB 20 (CA) in whichthe husband and wife were members of the Hasidic Jewish community, which factuallygenerated a relationship of complete trust and confidence between a wife and her husband inrelation to financial matters).

125 See, eg, WHD Winder, ‘Undue Influence and Fiduciary Relationship’ (1940) 4 TheConveyancer and Property Lawyer 274, to the effect that the law of undue influence does notapply to transactions for value, only to gifts—abuse of confidence is the proper ground forinterference with transactions for value; G Spencer Bower, The Law Relating to ActionableNon-Disclosure (London, Butterworths, 1915) 272–3, explaining that the two relationsco-exist and coalesce, but the duties that arise out of them are merely ‘accidental’ and must bedistinguished; cf G Spencer Bower, AK Turner and RJ Sutton, The Law Relating to ActionableNon-Disclosure, 2nd edn (London, Butterworths, 1990) §§ 16.07, 21.04–21.06, agreeingwith the statements in first edition of the work; RP Austin, ‘The Corporate Fiduciary:Standard Investments Ltd v Canadian Imperial Bank of Commerce’ (1986) 12 CanadianBusiness Law Journal 96, 97: ‘the two doctrines are conceptually different’; P Birks and NYChin, ‘On the Nature of Undue Influence’ in J Beatson and D Friedmann (eds), Good Faithand Fault in Contract Law (Oxford, Oxford University Press, 1995) 91: ‘It is less useful, evendangerous, to create a close relationship between undue influence and breach of fiduciaryduty’; S Worthington, above n 50, at 503, fn 18, arguing that the law of undue influence ‘iscompletely independent of and distinct from’ fiduciary law.

126 See, eg, PV Baker and P St J Langan, Snell’s Principles of Equity, 28th edn (London,Sweet & Maxwell, 1982) 544, cited with approval by Slade LJ in Bank of Credit andCommerce International SA v Aboody, above n 114, at 962 (although the House of Lords inPitt, above n 7, at 208, comments that the Court of Appeal in Aboody regarded the abuse ofconfidence (breach of fiduciary duty) line of cases as ‘a wholly separate doctrine of equity’);M Cope, Duress, Undue Influence and Unconscientious Bargains (Sydney, Law Book Co,1985) 79ff; RP Meagher, JD Heydon and MJ Leeming, Equity: Doctrines and Remedies, 4thedn (Sydney, Butterworths, 2002) 514.

127 [1911] 1 Ch 723 (CA) 728: ‘Fiduciary relations are of many types; they extend fromthe relation of myself to an errand boy who is bound to bring me back my change up to themost intimate and confidential relations which can possibly exist between one party andanother where the one is wholly in the hands of the other because of his infinite trust in him.’

128 Flannigan, ‘Commercial Fiduciary Obligation,’ above n 49, at 911, eg, complains thatwe often lose sight of the ubiquitous and generic nature of fiduciary obligation, which is a factthat should always inform our approach to fiduciary issues: ‘We lose track of this [fact], it

410 Rick Bigwood



fiduciary tasks or functions routinely arise within relationships that are not‘fiduciary’ for all intents and purposes.129 This observation underscores thepurely formal nature of the common assertion that the ‘fiduciary relation-ship’ and the ‘relationship of influence’ (for the purposes of a Class 2undue influence claim) overlap but are not identical. Consider, for instance,the following textbook example, which is very commonly encountered:

some relationships of influence are clearly not also fiduciary in character (forexample, parent and child) and it is clear that the categories are not co-incidentthough they have limited areas in common.130

But tasks or functions that happen to be performed by actors who are nottypically fiduciaries for all intents and purposes must nevertheless beregarded as ‘fiduciary’ to the extent that they involve the actor acquiringdirect or indirect access to another’s assets for the defined and limitedpurpose of the parties’ relationship or undertaking, rather than for theactor’s own, unencumbered purposes.131 So, for example, although theparent–child and guardian–ward relationships are not formally regarded as‘fiduciary relationships’ in toto, parents and guardians are nevertheless‘fiduciaries’ in virtue of, and to the extent of, the ‘tutelary’ and ‘advisory’functions they each perform in respect of their respective unemancipatedchildren and wards.132 Such functions give the functionary a peculiarability to influence the choices and actions of the beneficiary who, in thecontext of the specific physical arrangement between the parties, isdependent on the functionary performing his or her special tasks in aconscientious and disinterested manner. In other words, the special tasksbecome the source of the deferential trust (the limited-access relation of

seems, when we ask whether a particular relationship is fiduciary. It is preferable to askwhether there are fiduciary obligations arising from a specific physical arrangement. Amechanic is not generally considered to be in a fiduciary relationship with customers. Yet amechanic will have, along with certain contract and tort obligations, certain fiduciaryobligations. In that sense (and to that limited extent), the relationship is fiduciary’ (emphasisin original). See also Flannigan, ‘Fiduciary Mechanics’, above n 67.

129 Hence the label ‘fiduciary’ is inadequate per se to distinguish between those (fiduciary)relationships that will give rise to a presumption of undue influence and those that will not.Not all relationships that are ‘fiduciary’ also involve ‘influence’ of the kind required to attractrelational undue influence regulation. This point is well recognised in the case law: ReCoomber, above n 127, at 726–7 (Cozens-Hardy MR), 730 (Buckley LJ); Tufton v Sperni,above n 44, at 530 (Jenkins LJ); Cowen v Piggott (1989) 1 Qd R 41 (QSC) 44 (McPherson J,as quoted by Connelly J in the Full Court).

130 Meagher, Heydon and Leeming, above n 126, at 514.131 Flannigan, ‘Fiduciary Regulation of Sexual Exploitation,’ above n 49, at 307. See also

BH McPherson, ‘Fiduciaries: Who Are They?’ (1998) 72 Australian Law Journal 288, 288.132 Likewise of the doctor–patient relationship: see Breen v Williams, above n 55, at 83

(Brennan CJ), 92 (Dawson and Toohey JJ), 107–108 (Gaudron and McHugh JJ), 134–5(Gummow J). Patients repose trust in their doctors, who in turn owe confidentialityobligations and (are presumed to) acquire ascendancy over their patients for the purpose ofundue influence regulation, but they are not otherwise fiduciaries.




influence) that attracts a ‘fiduciary’ obligation for the purposes of Class 2relational undue influence law. Equally, guardians, doctors, spiritual advis-ers and the like, are ‘fiduciaries’ by virtue and to the extent of their abilityto influence their wards, patients and penitents (respectively) in theircapacity as ‘impartial advisers’ to such persons. Indeed, it has long beenrecognised that relationships can have varied purposes, some of which, orpoints within which, attract a ‘fiduciary’ obligation while others do not.133

Relational undue influence claims, which imply breach of deferential trust,are cardinal illustrations of this principle.

B. Royal Bank of Scotland plc v Etridge (No 2)

Morgan was not well received by all commentators and courts,134 at least inrespect of some of its points of law. In Goldsworthy v Brickell,135 for example,Nourse LJ refused to accept that, regarding the type of relation that attractedthe presumption of undue influence, the House of Lords ‘could have intended,sub silentio, to overrule not only Tufton v Sperni … but many other leadingcases from Huguenin v Baseley … onwards’.136 In Etridge, though, the Houseof Lords could have rescued the fiduciary account of relational undueinfluence but did not. Although one might have hoped that it would havedirectly and robustly examined the interaction, if any, between relationalundue influence and fiduciary obligation—as had, after all, been earliersignalled as a possibility by Lord Browne-Wilkinson in Pitt—that simply didnot occur.137 It seems clear that the House of Lords was not interested inreversing the trend set by Morgan, but the justification for that stance remainsfar from clear. There is in Etridge virtually no critical dissection of the earlier

133 New Zealand Netherlands Society ‘Oranje’ Inc v Kuys [1973] 2 All ER 1222 (PC)1225 (Lord Wilberforce); Hospital Products Ltd v United States Surgical Corporation, aboven 62, at 98 (Mason J); Noranda Australia Ltd v Lachlan Resources NL (1988) 14 NSWLR 1(SC) 15 (Bryson J): ‘a person under a fiduciary obligation to another should be under thatobligation in relation to a defined area of conduct, and exempt from the obligation in allother respects. Except in the defined area, a person under a fiduciary duty retains his owneconomic liberty’. This parallels closely courts’ treatment of the purchasing rule in fiduciarylaw. Thus, as Lord Blackburn observed in McPherson v Watt (1877) 3 App Cas 254 (HL)270–71, eg, a solicitor–client case, ‘If [the fiduciary] purchases … in a matter totallyunconnected with what he was employed in before, no doubt an attorney may purchase fromone who has been his client, just as any stranger may do … being in no respect bound to domore than any other purchaser would do. But when he is purchasing from a person propertywith respect to which the confidential relation has existed or exists, it becomes wrong of himto purchase without doing a great deal more than would be expected from a stranger’(emphasis added).

134 A convenient summary of the various academic views on Morgan, offered shortly afterthe decision, can be found in Geffen v Goodman Estate, above n 16, at 222–6 (Wilson J).

135 Above n 44.136 Ibid, at 406.137 Only Lord Clyde mentions Lord Browne-Wilkinson’s passage, but he does not respond

to its invitation: Etridge, above n 6, at [104].

412 Rick Bigwood



authorities on the subject, and much of the elaboration of the ‘first principles’of undue influence reads as assertion rather than analysis. The signals sentabout the nature of liability for relational undue influence are mixed, andthere is appreciable internal inconsistency in the individual judgments.138 As aresult, commentators on the case seem to have taken slightly differentmessages from it, as have later courts purporting to describe and apply thenow reigning law.139

By way of illustration of the doctrinal and conceptual instability createdby the exposition of undue influence in Etridge, consider first LordNicholls’ description of that form of ‘unacceptable conduct’ that ‘arises outof a relationship between two persons where one has acquired overanother a measure of influence, or ascendancy, of which the ascendantperson then takes unfair advantage’140—cases involving what we have beencalling ‘relational’ or ‘Class 2’ undue influence. He distinguishes thosecases from another form of unacceptable conduct (in the manner ofwrongful persuasion) that equity broadly identified as falling within theprinciples of undue influence, namely, ‘overt acts of improper pressure orcoercion such as unlawful threats’141—cases involving so-called ‘non-relational’ or ‘Class 1’ undue influence. In describing the nature of theinfluence that is abused in the ‘relationship’ cases, Lord Nicholls explainshow it ‘provides scope for misuse without any specific overt acts ofpersuasion’142 (hence its very nature makes monitoring and detectiondifficult). He continues, in language that is classically ‘fiduciary’ in itspurport, as demonstrated in particular by the following emphasised words:

The relationship between two individuals may be such that, without more, oneof them is disposed to agree a course of action proposed by the other. Typicallythis occurs when one person places trust in another to look after his affairs andinterests, and the latter betrays this trust by preferring his own interests. Heabuses the influence he has acquired. In Allcard v Skinner … Lindley LJ …described this class of cases as those in which it was the duty of one party toadvise the other or to manage his property for him. In Zamet v Hyman … LordEvershed MR referred to relationships where one party owed the other anobligation of candour and protection.143

138 Although all of the Law Lords purport to agree with Lord Nicholls’ judgment.139 See, eg, Hammond v Osborn [2002] EWCA Civ 885; Glanville v Glanville [2002]

EWHC 1587 (Ch); Niersmans v Presticcio [2004] EWCA Civ 372.140 Etridge, above n 6, at [8].141 Ibid. The reference here to ‘unlawful threats’ is immediately anomalous, as such

threats, if effectual, would clearly constitute duress at common law, whereas equity tended toregulate coercive conduct that fell below strict illegality such as lawful but unconscientiouspressure. Lord Hobhouse also stated that Class 1 undue influence ‘is capable of includingconduct which might give a defence at law, for example, duress and misrepresentation’: ibid,at [103].

142 Ibid, at [9].143 Ibid (emphasis added).




Lord Nicholls mentions ‘parent and child’ as an example of the type ofrelation in equitable contemplation here, but he rightly accepts that therelationships in which the Class 2 principles fall to be applied are incapableof being listed exhaustively.144

So far so good, at least for the fiduciary account, but in the very nextparagraph of his judgment Lord Nicholls potentially explodes the fiduciaryboundaries of the so-called ‘relationship’ cases:

Even this test [of whether ‘one party has reposed sufficient trust and confidencein the other’] is not comprehensive. The principle [applicable to the ‘relationship’cases] is not confined to cases of abuse of trust and confidence. It also includes,for instance, cases where a vulnerable person has been exploited.145

He then asserts that ‘no single touchstone [exists] … for determiningwhether the principle is applicable.’ He observes:

Several expressions have been used in an endeavour to encapsulate the essence:trust and confidence, reliance, dependence or vulnerability on the one hand andascendancy, domination or control on the other. None of these descriptions isperfect. None is all embracing. Each has its proper place.146

Here, there are problems with Lord Nicholls’ beliefs and analysis. First, byappearing to extend the scope of the relational undue influence principlebeyond limited-access relations of influence (deferential trust) to wider,arm’s-length power–vulnerability relations (exploitable disadvantage), hefails to observe the distinctive character of relational undue influencecircumscribed by the fiduciary principle, and indeed risks rendering Class 2undue influence conceptually and practically indistinguishable from mereunconscionable dealing, particularly when the exploitation involved is ofthe ‘passive’ kind, or by omission only.147 And although it is not clear thatit was Lord Nicholls’ intention to collapse relational undue influence intounconscionable dealing, for there is no explicit attempt anywhere inEtridge to reconcile undue influence with its equitable sibling, it isnonetheless a reasonable reading of the words that he chose to express the

144 Ibid, at [10].145 Ibid, at [11].146 Ibid.147 Not everyone considers such conceptual intermixture to be undesirable. See, eg, I

Hardingham, ‘Unconscionable Dealing’ in P Finn (ed), Essays in Equity (Sydney, Law BookCo, 1985) 18; A Phang, ‘Undue Influence—Methodology, Sources and Linkages’ [1995]Journal of Business Law 552; D Capper, ‘Undue Influence and Unconscionability: ARationalisation’ (1998) 114 LQR 479; A Phang and H Tjio, ‘The Uncertain Boundaries ofUndue Influence’ [2002] Lloyds Maritime and Commercial Law Quarterly 231, 232–4,241–3; A Phang and H Tjio, ‘Drawing Lines in the Sand? Duress, Undue Influence andUnconscionability Revisited’ (2003) 11 Restitution Law Review 110, 117ff. Cf also JDevenney and A Chandler, ‘Unconscionability and the Taxonomy of Undue Influence’ [2007]Journal of Business Law 541.

414 Rick Bigwood



jurisdiction and the natural effect of the approach to the burden andmethods of proof in connection with undue influence claims in general—ofwhich more shortly.

Moreover, the extension of relational undue influence to encompasswider power–vulnerability situations is simply asserted in Lord Nicholls’judgment. No supporting authority is mentioned and no justification isgiven. One is left instead to speculate. Certainly no example of relationalundue influence occurring outside a (subjectively?) ‘trusting and confiden-tial’ relationship is given at the relevant point in his judgment. It is likelythat he merely had in mind, and so was affirming sub silentio, a point thathad been made in the Court of Appeal below, where Stuart-Smith LJsuggested that undue influence was not confined to ‘abuse of confidence’principles.148 But neither of the examples that Stuart-Smith LJ cited as anexample of ‘undue influence’ by exploitation of a person who is vulnerableotherwise than by reason of being in a trusting and confidential relation-ship with her exploiter—Allcard v Skinner and Re Craig—are inconsistentwith the traditional fiduciary account of the Class 2 cases.149 As seen inJohnson v Buttress, for example, a limited-access relationship of influencecan equally result from known ‘ascendancy and dependence’ inter partes, ifthat ascendancy and dependence is of a ‘deferential’ kind, as from anactual concession of (deferential) trust and confidence in another. Indeed,earlier in his judgment Lord Nicholls himself cited Allcard v Skinner as atypical ‘trust’-type case,150 which is at variance with his apparent accept-ance of Stuart-Smith LJ’s observations in the Court of Appeal below. Atleast on one reading of Allcard v Skinner—and granted the judgmentsthere are open to diverse interpretations151—the lady superior in that caseenjoyed ‘limited access’ to the claimant’s assets, and hence was subject tofiduciary accountability, by virtue of the special ‘advisory’ function that ishabitually performed by spiritual leaders in relation to their devotees (not

148 [1998] 4 All ER 705, [8]: ‘The equitable doctrine of undue influence … is not confinedto cases of abuse of trust and confidence; it is also concerned to protect the vulnerable fromexploitation. It is brought into play whenever one party has acted unconscionably inexploiting the power to direct the conduct of another which is derived from the relationshipbetween them. This need not be a relationship of trust and confidence; it may be arelationship of ascendancy and dependence.’ Again, and with respect, this passage misses thepoint that it is ‘limited access’ regardless of its source (trust and confidence, ascendancy anddependence, or whatever) that attracts the Class 2 jurisdiction. The remainder of theparagraph in Stuart-Smith LJ’s judgment, however, confirms that he was talking of Class 1undue influence occurring within the context of ‘some close and confidential relation to thedonor,’ as he begins the sentence following the above-quoted words with: ‘In such cases actualundue influence has been said to involve’ (emphasis added).

149 Above n 7 and above n 6 respectively.150 Etridge, above n 6, at [9].151 As Charlotte Smith’s recent essay shows: C Smith, ‘Allcard v Skinner (1887)’ in C

Mitchell and P Mitchell (eds), Landmark Cases in the Law of Restitution (Oxford, HartPublishing, 2006) ch 8.




to mention by virtue of the particular rules of the sisterhood by which theclaimant donor had become bound). And in Re Craig, the companion/housekeeper became an ad hoc ‘fiduciary’ because of the elderly donor’sknown dependency on her ‘for his comforts and emotionally for hercompanionship and for her participation in his business affairs’, andbecause of her comparative powers, abilities, competency, managing dispo-sition, strong personality and physical and mental toughness.152 As a resultof these features of the parties’ respective positions, Ungoed-Thomas Jconcluded that the companion/housekeeper ‘had a duty in the circum-stances to advise [the elderly donor], and if not to advise him at any rate totake care of him, in the management and disposal of his property’.153 This,certainly, much like Johnson v Buttress before it, is consistent with a‘fiduciary’ function being imposed on the companion/housekeeper byvirtue of a limited-access relation of influence having resulted from thecumulative circumstances of the case, in particular because a fact-baseddeferentially trusting relationship had been established.

In the final analysis, Lord Nicholls’ suggestion that ‘no single touchstone[exists] for determining whether the principle [in the ‘relationship cases’] isapplicable’ is telling of the House of Lords’ failure to appreciate fully thefiduciary nature of all relational undue influence claims, and why that isso. There is, in fact, a single determinative yardstick for operation of theprinciple in the Class 2 cases—namely, whether the defendant’s access tothe plaintiff’s assets is qualified by a limited purpose—and none of theexpressions mentioned by Lord Nicholls that have been used to capture theessence of the cases—‘trust and confidence, reliance, dependence or vulner-ability on the one hand and ascendancy, domination or control on theother’—exist meaningfully or authoritatively in isolation of that inquiry.None of them alone expresses the essence of the requisite relation ofinfluence; all are merely proxies for ‘limited access to assets’ or elsedescriptions of various factual phenomena that might, severally or incombination, with or without the aid of presumption,154 lead a court todetermine that a limited-access relation of influence existed between theparties.155 The imposition of fiduciary obligation and regulation simplyfollows from that ultimate finding. Ironically, although Lord Nichollsstates that each expression ‘has its proper place,’ nowhere are we told whatthat is exactly. The answer, though, is that each expression is usefulbecause each, properly understood, performs an important signallingfunction as to the possible existence of a limited-access relation of

152 Above n 6, at 119–20 (Ungoed-Thomas J).153 Ibid, at 120.154 That is, in the Class 2(A) cases, where a presumption of limited-access is utilised.155 Which is really why, in Lord Nicholls’ words, none of the expressions is ‘perfect’ or ‘all

embracing’.

416 Rick Bigwood



influence inter se. That type of relation is in fact the ‘single touchstone’that the Law Lords in Etridge could have identified based on theconventional jurisprudence available to it before Morgan, but whoseobscurity it managed to perpetuate instead.

The other Law Lords in Etridge send equally mixed signals as to theessential nature of the relationship that attracts the principles of relationalundue influence. Lord Hobhouse, for example, describes the Class 2,‘presumed’ undue influence cases as ‘necessarily involv[ing] some legallyrecognised relationship between the two parties,’ the result of which ‘oneparty is treated as owing a special duty to deal fairly with the other. …Typically they are fiduciary or closely analogous relationships.’156 His useof the word ‘typically’ would indicate that other types of relationship areenvisioned here, but no clue is given as to what those are. Reference to therelationship being one in which one party owes a ‘special duty to dealfairly with the other’ is vague and begs the question entirely. Generalarm’s-length relationships can engender duties of ‘fairness’ (for examplenot to coerce or mislead), as can specific relationships that are not part ofrelational undue influence law such as relationships involving specialdisadvantage that attract the potential operation of the unconscionabledealing doctrine. Lord Hobhouse cites an example of a solicitor owing aduty to ‘deal fairly’ with a client when purchasing property from him orher,157 presumably within the scope of the solicitor’s fiduciary function,158

but then in turn fails to distinguish the regular breach of fiduciary dutysituation from relational undue influence. The latter involves a breach ofdeferential trust only (and, depending on the facts, the solicitor may havebeen trusted in a vigilant sense only, or in a deferential sense only, or inboth senses concurrently).159 Lord Hobhouse also fails to acknowledgethat the purchasing solicitor’s burden of demonstrating ‘fair dealing’ in theexample is not a manifestation of an obligation simply to ‘act fairly’ inrelation to business transactions with a client. Rather, it is part and parcelof the singular and generic ‘fiduciary’ responsibility to act ‘disinterestedly’toward him or her, in the sense that the fair-dealing rule that applies tofiduciaries who purchase from their obligees within the scope of thefiduciary relationship or function must be understood against the backdrop

156 Etridge, above n 6, at [104] (emphasis added).157 Ibid.158 In Allison v Clayhills (1907) 97 LT 709 (Ch) 712, Parker J suggested that the

presumption of undue influence would not apply if a solicitor bought a horse from a clientwho had retained him to conduct an action for slander, the relationship of influence existingonly in respect of the latter.

159 Clients tend to expect solicitors to act as their agents in implementing decisions (raisingthe intermediary cost mischief) and, in the same or another transaction, expect them also toadvise in a disinterested manner (raising the undue influence mischief). Indeed, Allcard vSkinner can be seen as a case where the particular spiritual adviser was trusted in both thevigilant and the deferential senses: see especially above n 7, at 182 (Lindley LJ).




of an overarching ‘no-conflict’ rule that does not apply to purchasingparties who might otherwise be duty-bound to ‘deal fairly’ with theircounterparty in the context of regular, arm’s-length transactional encoun-ters.

In his judgment Lord Scott speaks in terms of relationships of ‘trust andconfidence’160 but appears to assign them no special meaning so as toindicate that the notion of fiduciary trust and confidence is necessarilyintended thereby. For example, he mentions a wife having ‘trust andconfidence’ in her husband’s ability to make financial and businessdecisions, but of course this does not necessarily imply that ‘limited access’,hence ‘fiduciary’ obligation, is necessarily established inter se. Indeed, hedistinguishes the relation of husband and wife from other relationships‘generally of a fiduciary character, where, as a matter of policy, the lawrequires the dominant party to justify the righteousness of the trans-action’.161 This suggests that Lord Scott might regard the ‘presumption’ inthe Class 2(A) cases as something different than the ‘presumption’ in theClass 2(B) cases,162 which he saw as ‘doing no more than recognising thatevidence of the relationship between the dominant and subservient parties,coupled with whatever other evidence is for the time being available, maybe sufficient to justify a finding of undue influence on the balance ofprobabilities’.163 Although this of course is to ‘doubt the utility of the Class2(B) classification’,164 presumably on redundancy grounds,165 in the endanalysis the point is asserted rather than validated in his judgment.Traditionally, Class 2(A) and Class 2(B) undue influence differed only inthe manner of the claimant’s proving the existence of the special ‘influence’claimed to have been abused in the procurement or receipt of the impugnedbenefit, the existence of such influence being merely one ingredient in therelational undue influence claim, together with some sort of evidence as toits actual or possible abuse. The nature of the influence regulated in bothcases, however, was otherwise identical, and it was limited-access,‘fiduciary’-type, influence.166 The distinction between the two sub-classesof relational undue influence claim merely echoed the familiar distinctionbetween ‘status-based’ and ‘fact-based’ fiduciaries that subsists in the widerfiduciary law. It followed, too, that the nature of the ‘presumption’ in each

160 See, eg, Etridge, above n 6, at [158]–[159].161 Ibid, at [158] (emphasis added).162 This seems to have led some to assume that there are now three types of undue

influence: actual, presumed and relational (a species of actual undue influence). See possiblyFerris, above n 47. However, the House of Lords makes it quite clear that, at least from aforensic standpoint, ‘presumed’ undue influence must also count as a form of ‘actual’ undueinfluence.

163 Etridge, above n 6, at [161].164 Ibid.165 Cf Lord Hobhouse, ibid, at [107]; Lord Scott, ibid, at [161].166 Cf Goldsworthy v Brickell, above n 44 at 401 (Nourse LJ).

418 Rick Bigwood



class was identical, and it was a device of substantive policy rather thantrue fact-finding. That the circumstances of an individual case mightincidentally justify an inference of undue influence as having actuallyoccurred inter partes, independent of artificial presumption—which, itmight be noted, is a possibility in connection with status-based Class 2(A)relations of influence as well as fact-based Class 2(B) ones—ought not toobscure the overall fact that the presumption, while strictly superfluous inthe instant case, might nevertheless continue to operate for its original,policy-based reasons. Indeed, in Johnson v Buttress Dixon J was alive tothis point when he observed the occasional practice, even at that time, ofsome courts, when armed with ‘the presence of circumstances which mightbe regarded as presumptive proof of express influence’, and in connectionwith the Class 2(B) special relations of influence in particular, to treat thecase as if it ‘were not governed by the presumption but depended on aninference of fact’.167 But for Dixon J, the existence of circumstances thatmight give rise to the suspicion that ‘active circumvention’ had beenpractised168 was merely an incidental fact, and a ‘cause why cases whichreally illustrate the effect of a special relation of influence in raising apresumption of invalidity are often taken to decide that express influencewhich is undue should be inferred from the circumstances’.169

C. On Burdens and Proof: From Persuasive ‘Presumption’ to Mere‘Permissible Inference’

If the House of Lords’ conceptualisation of the relations or encounters thatattract the principles of relational undue influence are equivocal as to theirelemental fiduciary character, there is no doubt as to the evisceration of theconventional fiduciary rationale by the unambiguous demotion of thepresumption, in the Class 2 cases, to a mere permissible (as opposed to amandatory and controlling) inference. Drawing an analogy with thedoctrine of res ipsa loquitur at common law, three Law Lords in Etridgeconsidered the onus of proof to rest throughout on the complainant whowas alleging undue influence, and that the effect of the ‘presumption ofundue influence’ commonly said to arise in the Class 2 cases, involvingso-called relationships of ‘trust and confidence’, was merely ‘descriptive of

167 Above n 15, at 135. In making these remarks Dixon J referred to Scrutton LJ’sobservation in Lancashire Loans Ltd v Black [1934] 1 KB 380 (CA) 404 that common lawjudges were inclined in this area ‘to rely more on individual proof than on generalpresumption, while considering the nature of the relationship and the presence of independentadvice as important, though not essential, matters to be considered on the question whetherthe transaction in question can be supported’.

168 For example the presence of manifestly inadequate consideration in combination withthe special antecedent relation between the parties.

169 Johnson v Buttress, above n 15, at 136 (emphasis added).




a shift in the evidential onus on a question of fact’.170 In other words,‘presumption’ here means ‘permissible inference’ in the sense that thenatural probative weight of the basic facts proven by the claimant as partof the burden of production (‘relation of influence’ and ‘inexplicabletransaction’) is itself sufficient to discharge the claimant’s persuasiveburden in the absence of adequate counterproof by the defendant.171 Assoon as the defendant adduces countervailing evidence under his or herburden of production, however, it is a matter then of deciding whether theclaimant has succeeded in meeting the ultimate persuasive burden ‘on thebalance of probabilities’, that is, after the court has

drawn appropriate inferences of fact upon a balanced consideration of the wholeof the evidence at the end of a trial in which the burden of proof rested upon theplaintiff. The use, in the course of the trial, of the forensic tool of a shift in theevidential burden of proof should not be permitted to obscure the overallposition.172

The court thus treats the defendant’s burden as a mere ‘evidential’ onus ora ‘provisional burden’173—a mere tactical requirement to adduce inferen-tially inconsistent evidence to affect the weight of the plaintiff’sevidence—as opposed to a ‘persuasive’ or ‘legal’ burden that the defendantmust discharge to the appropriate standard of proof so as to avoidsummarily losing. The defendant is not, in other words, required to adduceevidence that would legally justify a finding against the presumed (inferred)fact.

The difference between what their Lordships in Etridge describe asoccurring forensically in relational undue influence claims and how DixonJ in Johnson v Buttress (for example)174 explains the operation of the

170 Above n 6, [16]. See also ibid, at [107], [161].171 Lord Nicholls, ibid, at [14]. Hence the analogy to res ipsa loquitur, where the basic

evidence is ‘of such a nature as to afford in itself sufficient proof of negligence’: Fitzpatrick vWalter E Cooper Pty Ltd (1935) 54 CLR 200 (HCA) 218 (Dixon J).

172 Etridge, above n 6, [16].173 Cf AT Denning, ‘Presumptions and Burdens’ (1945) 61 LQR 379, 380.174 I do not want to create the impression that the court’s view in Johnson v Buttress as to

the evidential effect of the presumption is universally accepted. It is frequently unclear in thecases whether the court, in applying the presumption of undue influence, is implementing arule concerning the burden of proof in the sense of the burden of persuasion or the burdenmerely of going forward with the evidence (the burden of production). One can readily locatedecisions to either effect, but seldom is the distinction between the two views discussed in thecase law. An exception (besides Etridge), which is contrary to the view that the legal burdenshifts in relational undue influence claims, can be found in Sopinka J’s judgment in Geffen vGoodman Estate, above n 16, at 242–3. Although Justice Sopinka accepted that thepresumption of undue influence was one of law, he noted the division among courts andcommentators as to the evidential effect of the presumption. His view, in obiter dicta, wasthat the matters that the defendant had to prove in rebutting the presumption of undueinfluence were merely factors that the court should consider in weighing the evidence and notan application of the legal burden of proof. In other words, no persuasive burden fell on the

420 Rick Bigwood



presumption in such cases under the conventional fiduciary rationale isobvious. Under the traditional fiduciary account, the presumption ofundue influence was a true legal presumption, with compelling effect,175

and not a mere ‘presumption of fact’ by which a court was permitted (butnot required) to deduce the ‘presumed’ fact, logically and reasonably, fromthe basic facts once they had been proven to the satisfaction of the court.The presumption did not cease to control as a makeweight in theassessment of the evidence unless the court positively believed the opposingevidence to the appropriate standard. The analogy to res ipsa loquitur inEtridge would, therefore, be inapt if relational undue influence were afiduciary rule, for the principle lacks the mandatory effect that distin-guishes true presumptions from mere permissible inferences.176 Moreover,as mentioned at the outset of this article, the force of the presumption inthe Class 2 cases was stronger than the basic facts needed to trigger it.Unlike the principle of res ipsa loquitur in common law negligence, thebasic facts, if unchallenged, did not have to preponderate of undueinfluence on their own. Indeed, to the extent that res ipsa loquitur can beconsidered a ‘presumption’ at all (so ignoring, arguendo, its lack ofmandatory effect), it is a very different sort of presumption, having a verydifferent legal effect, than the one traditionally associated with ‘presumed’undue influence. Res ipsa loquitur is a ‘probability presumption’,177 in thesense that it arises from ‘human experience or tendency’ and is based on

defendant in an undue influence claim; rather, his or her burden was merely an evidential one,albeit an evidential burden that was subject to ‘the familiar maxim that in applying thestandard of proof all evidence is to be weighed in light of the gravity of the issue to bedecided’: ibid, at 243, citing Blyth v Blyth [1996] AC 643 (HL) 673 and Hornal v NeubergerProducts Ltd [1957] 1 QB 247 (CA) 266). Thus, the evidence in undue influence cases ‘issubjected to close scrutiny and the cases indicate that before coming to a conclusion, the courtmust act only on clear and convincing evidence’: ibid, at 243. The debate is a difficult one thatcannot be resolved within the confines of this article. Indeed, many experts on evidence lawvigorously contend that it is inaccurate in any circumstance to speak of a shifting of the legalburden (on a single issue) in the course of a trial. Others, however, suggest that it is arguablethat certain presumptions do have this effect. See JD Heydon (ed), Cross on Evidence, 7th edn(Sydney, Butterworths, 2004) [7225], [7300]–[7310].

175 That is to say, success on the basic facts compelled the operation of the presumption asa matter of law.

176 ‘The principle expressed in the phrase res ipsa loquitur does no more than furnish apresumption of fact’: Fitzpatrick v Walter E Cooper Pty Ltd, above n 171, at 219 (Dixon J).It is merely the application of inferential reasoning and does not, like true presumptions, havemandatory effect. Cf Schellenberg v Tunnel Holdings Pty Ltd (2000) 200 CLR 121 (HCA)[22] (Gleeson CJ and McHugh J). I have in the past described the House of Lords’ analogy tores ipsa loquitur in this context as a ‘fair but limited one’: Bigwood, above n 17, at 439, fn 28;Bigwood, above n 40, at 385–86, fn 69. On further reflection, I no longer consider it to befair at all.

177 Here I draw on LJ Cohen, ‘Presumptions According to Purpose: A FunctionalApproach’ (1981) 45 Albany Law Review 1079, 1092–93, who argues that presumptions canbe classified into three major types, according to the purpose for which each was created:‘procedural presumption’, ‘probability presumption’, and ‘policy presumption’. Needless to




the strong practical likelihood or high probability that if the basic facts areshown, the presumed fact also exists.

The presumption of undue influence, in contrast, has traditionallyfunctioned as a ‘policy presumption,’ in the sense that it results not simplyfrom human experience or tendency as a matter of realistic possibility (letalone probability),178 but rather, and more significantly, from ‘societalpredilections toward the existence of the presumed fact’.179 As a policypresumption reflecting societal values, therefore, the presumption of undueinfluence received, because it deserved, a higher priority than thosepresumptions that merely reflected common sense, experience or humantendencies (such as res ipsa loquitur). As the strongest possible form ofpresumption it thus functioned, again unlike res ipsa loquitur, to shift theburden of proof (or the risk of non-persuasion), and not merely theevidential burden or burden of production, onto the other party. Withpolicy presumptions, the court must always find the presumed fact unlesspersuaded by a preponderance of the evidence, or, in the fiduciary context,high-probability (‘clear and convincing,’ ‘cogent and compelling’) proof,180

that the presumed fact does not exist.181 The presumption’s purpose thusdetermined the treatment it received in terms of: (a) the cogency of thebasic facts that triggered the presumption (basic facts that disclosed arealistic possibility of abuse as opposed to the substantial likelihood of it);(b) the presumption’s legal effect in creating a persuasive burden upon thedefendant to supply the want of necessary evidence; and (c) the standard ofproof that the defendant then had to meet in supplying that necessaryevidence (‘clear and convincing’, ‘cogent and compelling’, as opposed tomerely ‘preponderating’ proof).

Returning to Etridge, the analogy of the presumption of undue influenceto res ipsa loquitur is simply asserted by the House of Lords. Not only is its‘inexplicability’ test too high in terms of the cogency of the evidence, or thelevel of risk, considered necessary to activate the presumption in the Class2 cases (concerning, in particular, what counts as a ‘suspicious transaction’

say, not many presumptions fall neatly within one class, for many are founded, at least inpart, on all three bases: efficiency or expedience, quantifiable human experience or tenden-cies, and social predilections or values.

178 As Cohen, ibid, at 1093 points out, policy presumptions ‘differ from probabilitypresumptions because the relationship between the basic and presumed facts cannot bequantified’.

179 Ibid.180 The counterproof must compel a finding against the presumed fact, so that ‘in-

between’ evidence will not suffice. Higher standards of proof are required when particularlyimportant individual interests are at stake, as demonstrated, eg, in criminal (and indeedfiduciary) law. Cf C Manolakas, ‘The Presumption of Undue Influence Resurrected: HeSaid/She Said is Back’ (2006) 37 McGeorge Law Review 33, 40.

181 In other words, the defendant must overcome the presumption, not merely meet orbalance it with counterproof so as to then leave it as a matter for the tribunal of fact todecide.

422 Rick Bigwood



in the context of a special relation of influence),182 the Law Lords omit toconsider the various purposes that underlie different presumptions, whereone size clearly does not fit all.183 Presumptions are created to accomplishmany different objectives,184 so in truth ‘[n]o general rule can … be laiddown as to the effect of a particular presumption in the actual trial of acase, for this depends upon the purpose it is designed to serve’.185 Certainlythere is no direct reference in the various Etridge judgments to the genericpolicy foundations of the presumption of undue influence, and so noattempt is made by any of the Law Lords to demote the traditionalpresumption on rational grounds (for example by showing that the policyconsiderations that originally justified the equitable device are obsolete,harsh, impractical or unjust).

Some Law Lords clearly had in mind the Class 2(B) cases when makingthe analogy to res ipsa loquitur, rather than the Class 2(A) cases, whichwere strangely singled out for separate treatment by Lord Nicholls andLord Scott in particular. I say ‘strangely’ here because, historically, fromthe standpoint of the presumption of undue influence, both types ofcase—Class 2(A) and 2(B)—functioned in the same way. The remarks ofLord Nicholls and Lord Scott in that regard, however, are confused (or atleast confusing), which again casts doubt on the extent to which the priorlaw, which they were abandoning sub silentio, was appreciated, at least inlight of the authorities as they stood in the United Kingdom before 1985.Consider, for example, the following passage from Lord Nicholls’ judg-ment:

The evidential presumption discussed above [in the ‘relationship’ cases] is to bedistinguished sharply from a different form of presumption which arises in somecases. The law has adopted a sternly protective attitude towards certain types ofrelationship in which one party acquires influence over another who is vulner-able and dependent and where, moreover, substantial gifts by the influenced or

182 To be sure, Lord Nicholls in Etridge, above n 6, [25], [29], [30] favoured LordScarman’s strict suspicion of abuse test in Morgan, above n 30, at 709—namely, whether theimpugned transaction was ‘explicable only on the basis that undue influence has beenexercised to procure it’ (emphasis added)—which is a departure from the fiduciary account tothe extent that it appears to implement ‘probability’ regulation rather than ‘possibility’regulation. For the difference between the two forms of risk regulation, see n 58 above.

183 Cohen, above n 177, at 1091: ‘Because presumptions are formulated to achievedifferent purposes, it makes little sense to adhere to a rigid, one-rule treatment policy for allpresumptions.’ See also K Broun, ‘The Unfulfillable Promise of One Rule for All Presump-tions’ (1984) 62 North Carolina Law Review 697.

184 Edmund Morgan once described the ‘most intellectually satisfying solution’ in the fieldof presumptions thus: ‘The effect of the establishment of the basic fact of a presumptionshould depend upon the weight of the reasons which caused its origin or justify its persistence.The reasons which call presumptions into being and the purposes they are designed to achieveare various’: EM Morgan, ‘Further Observations on Presumptions’ (1943) 16 SouthernCalifornia Law Review 245, 250–51.

185 O’Dea v Amodeo 170 A 486, 487 (Conn Sup Ct Err 1934).




vulnerable person are not normally to be expected. Examples of relationshipswithin this special class are parent and child, guardian and ward, trustee andbeneficiary, solicitor and client, and medical adviser and patient. In these casesthe law presumes, irrebuttably, that one party had influence over the other. Thecomplainant need not prove he actually reposed trust and confidence in the otherparty. It is sufficient for him to prove the existence of the type of relationship.186

There are difficulties with this passage, which to my mind cast doubt onthe soundness of Lord Nicholls’ understanding of the Class 2(A) cases andtheir (traditional) relationship to the Class 2(B) cases. Although it isrecognised, correctly, that there are dual presumptions operating in theClass 2(A) cases—first a presumption of influence based on proof of astatus-based relation (solicitor–client, guardian–ward, and the like), andsecond a presumption of undue influence triggered by proof of a trans-action calling for explanation within the context of that relation—and thatthose two presumptions are to be ‘distinguished sharply’ from each other,Lord Nicholls seems to transpose the true marks of distinction. Hesuggests, on my reading of the above passage, that the second presumptionis identical to the ‘evidential presumption discussed above’ (a permissibleinference only), and that the first presumption somehow arises as a resultof ‘a sternly protective attitude towards certain types of relationship inwhich one party acquires influence over another who is vulnerable anddependent’ (albeit in combination with an inexplicable transaction). Infact, as demonstrated by Dixon J’s judgment in Johnson v Buttress, thelaw’s ‘sternly protective attitude’ explains entirely the sequent presumptionof undue influence in the relationship cases, both Class 2(A) and 2(B), andis not demonstrated at all in and by the initial presumption merely ofinfluence in the status-based Class 2(A) cases.

In terms of the presumption types discussed above, the first presumptionis at best a common-experience ‘probability presumption’ (a provisionalpresumption of fact),187 used to satisfy the legal burden on the claimant toprove merely one of the basic facts, whereas the latter is a much stronger‘policy presumption’ (a compelling presumption of law) serving a verydifferent function and having an entirely different legal effect.188 That LordNicholls suggests that the presumption of the first type is ‘irrebuttable,’allowing the court no freedom to draw a different conclusion than the onerequired, regardless of the strength of the opposing evidence as to the truth

186 Etridge, above n 6, at [18].187 Clarke v Hawke (1865) 11 Gr 527, 544.188 This is exemplified by the fact that as social views of relationship-types change, so will

the presumption. In other words, status-based relations of influence are very likely to contractin proportion to the changing roles and capacities of the parties to those relations in fact. See,eg, Lord Evershed MR’s remarks in relation to ‘the position of women in modern society’ inZamet v Hyman [1961] 3 All ER 933 (CA) 938 (concerning engaged couples).

424 Rick Bigwood



of the matter, defies both reason189 and prior authority.190 It also contra-dicts his express approval earlier in his judgment191 of Sir Guenter Treitel’sdefensible remark192 that the question in the ‘relationship’ cases is whetherone ‘party has reposed sufficient trust and confidence in the other, ratherthan whether the relationship between the parties belongs to a particulartype’.193 With respect, this is all symptomatic of a discomforting instabilitythat infects much of the doctrinal elaboration in the various Etridgejudgments, including that which eventually leads to an implicit, but I thinkinevitable, abandonment of the fiduciary approach to relational undueinfluence.

As mentioned above, Lord Scott thought that the Class 2(A) cases wereuseful in identifying particular relations where the presumption arose, buthe doubted the utility of the Class 2(B) classification.194 He viewed theClass 2(B) cases as examples of cases where, in proving the relationshipbetween the dominant and subservient parties, in combination withwhatever other evidence could be mustered at that time, the claimant hadafforded sufficient primary evidence to justify a finding of undue influencein the absence of contrary evidence on the other side—hence the analogy tothe res ipsa loquitur principle at common law.195 But as was alsomentioned, and as Dixon J in Johnson v Buttress explained, this does notmean that such cases are not otherwise capable of being based on the true,

189 First, the idea of a presumption being ‘irrebuttable’ means that it is not really a‘presumption’ at all; rather, it can only be a substantive principle expressed in the language ofa presumption. Secondly, what reason could exist for inferring, conclusively, interpersonalinfluence when such influence patently did not exist in fact? Cf Finn, above n 47, at 85.

190 Courts have, in fact, admitted evidence to controvert the presumption of influence inthe Class 2(A) category of case. See, eg, Westmelton (Vic) Pty Ltd v Archer [1982] VR 305(SC) where the presumption of influence was rebutted in a solicitor–client case. Cf alsoGoldsworthy v Brickell, above n 44, at 401 (Nourse LJ) holding that influence is presumed inthe Class 2(A) relationships ‘unless the contrary is proved’; Geffen v Goodman Estate, aboven 16, at 221 (Wilson J): ‘Equity has recognized that transactions between persons standing incertain relationships with one another will be presumed to be relationships of influence untilthe contrary is shown’ (emphasis added). Lord Nicholls in Etridge seems simply to blindlyadopt a point made per incuriam by Stuart-Smith LJ in the Court below, above n 148, at [6].

191 Etridge, above n 6, [10].192 I have also argued that the Class 2(A) presumption of influence should be abandoned

in favour of proof of an actual relation of influence in each case, while retaining the policyfoundations of the presumption of undue influence that follows upon proof of a suspicioustransaction occurring within the scope of the factual relation of influence: see Bigwood, aboven 40, at 433–4. In the wider fiduciary context, Flannigan has argued that, in identifyingfiduciary status, the formal relationship ought at best to perform a ‘modest signaling function’only, and that it ‘may be preferable to detach ourselves from our remaining dependence on thestatus ascription of fiduciary responsibility, and move to a fact-based limited access test for allcases’: Flannigan, above n 48, at 228–9. See also R Flannigan, ‘The [Fiduciary] Duty ofFidelity’, above n 49, text following fn 76.

193 GH Treitel, The Law of Contract, 10th edn (London, Sweet & Maxwell, 1999)380–81.

194 Etridge, above n 6, [161].195 Ibid.




policy-inspired presumption. Rare will be the case where a claimant reliesexclusively on the presumption for his or her proof, submitting no otherevidence. A party who claims to have been victimised by relational undueinfluence will naturally attempt to adduce as much evidence as possible tosubstantiate the claim, so seldom (if ever) will a court be asked to decidesuch a claim on a set of facts that presented nothing but the bare relationand the impugned transaction, and excluded all other evidentiary facts andcircumstances of probative inferential worth. This is true of both the Class2(A) and the Class 2(B) claims. Often the claimant will be able to carry hisor her burden of production so well in proving a Class 2(B) relation ofinfluence and contextually suspicious transaction that the evidence goessignificantly beyond the basic evidentiary facts that ordinarily would havesufficed to trigger the policy-based presumption. Often such basic facts willthemselves supply sufficient, but not necessarily conclusive, evidence of anabuse of special influence in fact (if not of some independent exculpatoryreason such as duress or unconscionable dealing), so as to support aninference of undue influence as a matter of logic and reasoning, independ-ent of invocation of the artificial presumption, thus allowing the claimantto make out a ‘prima facie case’ of undue influence and to meet his or herburden of persuasion, without the presumption, if there is too littleinferentially inconsistent evidence produced on the other side. In such casesthe same result can be reached without the aid of the presumption, makingits operation strictly redundant, but these cases do not remove thejustification for the presumption, or give reason to deny its operation in theparticular case; they merely remove the practical need for it in theparticular case (and perhaps in most, but certainly not all,196 Class 2(B)cases).197 The presumption continues to function, albeit merely theoreti-cally and superfluously, in all such cases.198 To my knowledge, the criteriafor activation of the policy-based presumption have never been defined in a

196 It is unlikely, eg, that Johnson v Buttress would have been decided in favour of thecomplainant under the Etridge evidential approach, unless one accepts Starke J’s view of thefacts in that case. See n 85 above and accompanying text.

197 See CB Mueller and LC Kirkpatrick, Evidence, 2nd edn (New York, Aspen Law &Business, 1999) § 3.4: ‘In effect, this kind of inference is the residue of a presumption, and[the tribunal of fact may] draw the conclusion that the presumption would otherwise require.’

198 Perhaps, therefore, ‘redundancy’ might be a ground for supporting demotion of thepresumption and elimination of the Class 2(B) category (as argued, eg, in a note, ‘UndueInfluence in Intervivos Transactions’ (1941) 41 Columbia Law Review 707), but theirLordships in Etridge do not make this argument at all, at least directly. Anyway, unless onethinks that Johnson v Buttress was wrongly decided, that case tends to demonstrate that theClass 2(B) category is not always redundant, as Johnson v Buttress presumably could nothave been decided the same way under the Etridge forensic approach. The majority of thecourt in Johnson v Buttress admitted that there was insufficient evidence to establish undueinfluence as a positive fact: see n 86 above. As mentioned in the text above, however, theviews expressed in Etridge about the disutility of the Class 2(B) category are asserted ratherthan fully reasoned.

426 Rick Bigwood



way that excludes the possibility of concurrent operation of a permissibleinference or probability-based presumption if the primary evidence sup-ports it.

IV. CONCLUSION

Contract lawyers are accustomed to courts killing off, or at least down-playing, ‘presumptions’ in their field.199 As McHugh J observed in Com-monwealth v Amann Aviation Pty Ltd, a reliance damages case:

the history of the law of evidence has seen an increasing rejection of presump-tions and other artificial forms of reasoning in favour of allowing tribunals offact to give such probative force to evidentiary materials as they think fit havingregard to all the circumstances of the case.200

It seems that the House of Lords in Etridge, for the United Kingdom atleast, has now consigned relational undue influence to that fate. Althoughthey do not explicitly reject the fiduciary character of (at least some of) therelations that attract the equitable jurisdiction, the Law Lords are unam-biguous in their view of the so-called ‘presumption’ of undue influence,and how it operates in the Class 2 line of cases. That view is incompatiblewith relational undue influence being maintained as a category within theframework of fiduciary regulation. Disappointingly, that view was assertedrather than fully reasoned. The conventional fiduciary rationale was notopenly considered and then rationally rejected. The traditional jurispru-dence appears simply to have been ignored. Assertion somehow becamethe substitute for analysis. The rejection was sub silentio. There wascertainly no frontal and convincing attack on the judicial intuition orsubstantive policy that originally justified strict fiduciary regulation inrelational undue influence cases.

That said, relaxation or abandonment of relational undue influence as afiduciary rule might be defensible on rational grounds—for example, becausethe rule can be shown to be obsolete, superfluous or impractical;201 or

199 Notably in the area of intention to create legal relations. See, eg, Fleming v Beevers[1994] 1 NZLR 385 (CA); Ermogenous v Greek Orthodox Community of SA Inc (2002) 187ALR 92 (HCA) [26] (Gaudron J, McHugh, Hayne, and Callinan JJ).

200 (1991) 174 CLR 64 (HCA) 166. This observation was in connection with thesuggestion, accepted in the court below, that the law presumed that expenditure in performinga contract would be recouped through such performance, so that a court was entitled to inferthat the plaintiff’s loss for the purpose of the assessment of damages was no less than theamount of that expenditure.

201 See, eg, the note at (1941) 41 Columbia Law Review 707, 722: ‘the relationships reliedupon to raise the presumption are essentially ambiguous, pointing at least as readily to a validreason for the favoritism exhibited in the particular transaction as to the conclusion thatsomething underhanded has taken place. It is therefore difficult to see any justification for theraising of a presumption of “undue influence”’.




‘paternalistic’;202 or to operate unfairly, harshly or unjustly;203 or to deprivetrial judges of the valuable ability to assess and weigh the evidence so as to getthe truth of the matter;204 or to be an ‘incubator in which fraud may flourishand grow’;205 or to create too-ready incentives for disgruntled or disap-pointed claimants to challenge objectively concluded transactions, oftenmany years after the event or the death of the transferor.206 Also, the Etridgeview of undue influence is not inconsistent with the (sometimes) consideredapproaches to Class 2 undue influence of several courts in other common lawlegal systems.207 Still, no such reasons are explicit in the demotion of thetraditional ‘presumption’ of undue influence in Etridge. The law in thatrespect is presented there as if it had always been that way.

It is hoped that other Commonwealth appellate courts,208 if and whenthey eventually confront the question of adopting (or not) the Etridgeundue influence principles,209 will respond reflectively, and not merelyreflexively, to the issue. Such courts should not, in my view, relax ordismantle the strict fiduciary operation of traditional relational undueinfluence law without first credibly responding to the conventional ration-ale that existed for the strict accountability regime in first place, that is,without furnishing plausible reasons why we should now comfortablyreject the original premises for the strict regulation. Without gainsaying thepossibility of such reasons coming forward, I would myself be surprised ifthe risk of, and hence judicial concern for, the mischief of opportunism inlimited-access arrangements were any less demonstrable today than in thepast. Surely those who concede deferential trust in others, or who areotherwise subordinate within a limited-access relation of influence, are as

202 CP Reed, ‘Comment’ (1984) 18 Law Teacher 132, 134.203 See, eg, BD Stapleton, ‘The Presumption of Undue Influence’ (1967) 17 University of

New Brunswick Law Journal 46, 64–5; A Craig, ‘Evidential Presumptions’ (2002) 152 NewLaw Journal 217, 218. Although fiduciary rules may operate harshly in some cases, this isoffset by the fact that it is very easy for a fiduciary to insulate himself or herself, ex ante,against that risk, that is, by securing the obligee’s permissive and transactional consent. Afailure to do so might itself be indicative, at least of transactional neglect. On the concept of‘transactional neglect,’ see R Bigwood, ‘Contracts by Unfair Advantage: From Exploitation toTransactional Neglect’ (2005) 25 OJLS 65.

204 Re Estate of Carpenter 253 So 2d 697, 703–704 (Fla 1975).205 Stapleton, above n 203, at 64.206 See, eg, the note at (1941) 41 Columbia Law Review 707, 722: ‘It seems, too, that

instead of being the shield of a trusting, weakminded grantor, as it was intended to be, the“presumption” can easily become the sword of a disgruntled, disappointed heir.’

207 For example, Boardman v Lorentzen 145 NW 750 (Wis 1914); Re Estate ofCarpenter, above n 204, a case involving the execution of a will, but the principles were laterheld to apply equally to inter vivos gift transactions; Majorana v Constantine 318 So 2d 185(Fla Ct App 2d Dist 1975); Re Estate of Wood, 132 NW 2d 35 (Mich 1965); Geffen vGoodman Estate, above n 16, at 242–3 (Sopinka J in obiter dicta).

208 Though perhaps Australia, Canada, and New Zealand especially.209 In contrast to the separate principles relating to third-party disability in the enforce-

ment of suretyship transactions, which is the main subject of the Etridge appeals.

428 Rick Bigwood



vulnerable to exploitation today as they ever were. On the other side, it isdoubtful whether twenty-first-century human beings, when tempted byready opportunities to make secret personal gains, are generally any lesslikely to succumb to the self-regarding impulse than their nineteenth-century counterparts. Indeed, in the wider fiduciary setting it has beenargued that ‘there does not appear to be a plausible basis for sending a newsignal that the regulation of opportunism [in limited-access arrangements]should now be scaled back’.210 Similarly, for undue influence, it might beargued that allowing the presumption of undue influence to affect only theburden of production and not the burden of persuasion is problematic onthe ground that the presumption itself embodies strong policy preferencesthat are not adequately served if only the burden of production is affected.

The ultimate effect of Morgan and Etridge, undoubtedly, is to enlargethe scope of undue influence so that it now potentially applies to all acts ofunfair persuasion in inter vivos transactions, regardless of the relationalcontext within which that persuasion occurs. To my mind this has renderedundue influence hollow as an independent doctrinal category. Stripped ofits fiduciary underpinnings, relational undue influence is now practicallyand intellectually indistinguishable from other exculpatory categories orpleas in avoidance that have, in other legal systems, tended to regulateunfair persuasion or victimisation in arm’s-length transactional encounters.Those categories or pleas are certainly capacious enough to absorb undueinfluence’s historical burden. In those other legal systems the expansion ofthe duress and unconscionable dealing doctrines, for example, has tendedto eliminate the need for a broader application of the undue influenceconcept, so it is possible that relational undue influence may, despiteEtridge, continue to retain its distinctive character in some English-basedlegal systems outside of the United Kingdom. The possible elimination ofthe Class 2(B) category of undue influence in particular, though, seems toleave little room for an understanding of relational undue influence apartfrom unconscionable dealing, the criteria of which seem perfectly wellsuited to the administration of many undue influence claims.211 Yet to theextent that both doctrines can be understood as legal devices for the

210 Flannigan, above n 54, at 212.211 As Mason J stated in a leading unconscionable dealing case (Commercial Bank of

Australia Ltd v Amadio, above n 11, at 461): ‘There is no reason for thinking that the tworemedies [of unconscionable dealing and undue influence] are mutually exclusive in the sensethat only one of them is available in a particular situation to the exclusion of the other. Reliefon the ground of unconscionable [dealing] will be granted when unconscientious advantage istaken of an innocent party whose will is overborne so that it is not independent andvoluntary, just as it will be granted when such advantage is taken of an innocent party who,though not deprived of an independent and voluntary will, is unable to make a worthwhilejudgment as to what is in his best interest.’




protection of vulnerable parties, it cannot be said that victims of uncon-scionable dealing and victims of relational undue influence are ‘vulnerable’(to victimisation) in quite the same way. ‘Specially disadvantaged’ partiesunder the unconscionable dealing jurisdiction are typically inept, weak orunable to advance their best interests when entering into voluntary orconsensual transactions with others who are much stronger, but that is notbecause they have renounced playing for advantage themselves in suchtransactions. They would or might press for advantage or pursue self-interest inter se if only they could, but ex hypothesi their special relativedisability inhibits them from doing so on the occasion in question. Suchpersons are disadvantaged, but they are not ‘exposed’. Deferentiallytrusting parties, in contrast, characteristically have surrendered, partiallyor completely, control in their decision-making to another and so aresusceptible to a much greater extent and to a higher order of wrongdoingaltogether. They are, to the extent of their surrender, truly exposed, andtheir exposure is to no less than betrayal or treachery. The corollary ofsuch greater vulnerability and risk on the one side must be greaterobligation on the other, which is fiduciary obligation. We may question,therefore, whether such parties would be adequately protected—theirinterests properly served—if their petitions for exculpation from impugnedtransactions were consigned to administration through what is effectivelyan unconscionable dealing inquiry only. That, I believe, is the ultimate andlamentable effect of Morgan and Etridge.212

212 Interestingly, in Lawrence v Poorah [2008] UKPC 21, Lord Walker, on behalf of theJudicial Committee, observed obiter dicta: ‘It is sufficient to say that the doctrines of undueinfluence and unconscionable bargain share a common root—equity’s concern to protect thevulnerable from economic harm—but they are generally regarded as distinct doctrines … Inparticular, although the doctrine of unconscionable bargain involves the exploitation of theplaintiff’s vulnerability, it does not depend on a pre-existing relation of actual or presumedconfidence. The doctrine of unconscionable bargain appears to be particularly vigorous inAustralian jurisprudence …’ (emphasis added).

430 Rick Bigwood



Index

Introductory Note

References such as ‘178–9’ indicate (not necessarily continuous) discussion of atopic across a range of pages. Wherever possible in the case of topics with manyreferences, these have either been divided into sub-topics or only the mostsignificant discussions of the topic are listed. Because the entire volume is about thelaw of contract, the use of this term (and certain others occurring throughout thework) as an entry point has been minimized. Information will be found under thecorresponding detailed topics.

abuse:of confidence, 406, 410, 415of rights, 92of trust and confidence, 388, 414–15

acceptancesee offer and acceptance

access to assets, 391, 393–5, 397–8, 402,405, 409, 416

accessory liability, 225, 229accommodationism, 115–21, 123, 128,

132–5, 137accuracy, 237–8, 243, 249, 252, 259–61,

263–4, 365acquiescence, 71, 73, 213, 215–17, 352action:

detrimental, 219–20, 224private rights of, 116, 128–33, 136–7,

306actor objectivity, 349–51, 364–5actual undue influence, 382, 399, 415, 418Addison, C, 51–2, 74agency, 304, 395–6agency costs, 396agents, 66, 113, 129, 242–3, 259–62, 395

moral, 115–21, 127–8, 130–2, 134–5,137

agreed exchange of performances, 330agreements, 37, 104–5, 150–2, 335–9,

355–6, 363–4, 372–8absolving an actor from liability for

fraud, 244, 246, 253binding, 120, 216contractual, 11, 210, 268, 290objective, 356, 359–60, 362, 364

alterations, 98, 277–8ambiguity, 341, 343, 354

latent, 363analogical reasoning, 122

anti-assignment clauses, 268, 279, 281,283–9, 291, 305–18

breaches, 309–17effective, 268, 288

Aristotle, 47, 370arm’s-length transactions, 388–9arrangements:

limited-access, 386, 393, 398, 401–2,428–9

physical, 391, 402, 411‘as is’ sales, 242–4, 254, 256, 357ascendancy, 381, 395–6, 399–400, 402–3,

411, 413–16assets, 63–4, 66, 145–7, 159–62, 164–6,

271–2, 393–5non-profit-making, 174

assignees, 269–77, 279–80, 282, 284–316,318

equitable, 270, 274, 291, 293, 307rights, 277, 279

assignment, 267–91, 293–318, 402of choses in action, 286, 289–90of contract rights, 267, 295, 304, 314contracts of, 286–8, 291, 316equitable, see equitable assignmentsinformal, 286–7of leases, 313–14purported, 283, 310, 316statutory, 270, 278, 283, 289, 291, 293,

301surrogate, 277

assumpsit, 53, 64assumptions, 116, 215–17, 252–3, 374

mistaken, 368–76relevant, 204–6, 209, 216–17, 219–20and terms, 371–3

assurances, 57, 211, 213, 220, 250, 259,308–9

Atiyah, PS, 43–4, 46

Columns Design Ltd / Job: Neyers / Division: Index /Pg. Position: 1 / Date: 16/4


auctioneers, 47, 362, 366Austin, J, 233authorisation, 386, 392autonomy, 32, 133, 347, 364

of the will, 38, 86–8

bare trusts, 271–2, 274–5, 277–8Beatson, J, 343beneficial reliance, 208benefits, impugned, 383, 405, 418Benson, P, 56, 121, 126, 207, 210, 213Bentham, J, 31best interests, 429–30BGB, 36–38bidders, 362, 366bindingness, 123, 127–8Birks, P, 26–9, 31, 71, 201–2, 210, 215Blackstone, W, 29–30, 52–4, 129–30boilerplate, 240–1, 247, 257breach, 139–42, 164–9, 174–6, 186–7,

226–33, 309–13, 315–18of covenant, 313–15of deferential trust, 404, 412, 417efficient, 16, 145of fiduciary duty, 275, 390, 410of promise, 47, 121, 127, 132, 207responses to, 142–4time of, 176, 182–3, 187, 191, 196of trust, 274–5, 404of warranty, 242, 254

building contractors, 283–4, 287–8, 316building contracts, 188, 283–4, 309,

316–17burden of proof, 384, 389, 401, 420–2

and relational undue influence, 419–27Burrows, A, 27, 71, 77, 191–4buyers, 176–83, 242–4, 252–4, 256–8,

263–5, 341, 356–63conduct, 358identity, 104intention, 357, 372mistake, 342, 357–8, 368

capacity, 19–21, 38, 53, 84, 194, 259, 265special, 382, 398–9

carelessness, 173, 208–9, 364, 378carriage of goods, 111, 179, 184–6carriers, 66, 111cartography of obligations, 25–49categories, 2–3, 17–18, 28–9, 201–2,

389–90, 425–7hardening of, 31, 38, 42, 48–9legal, 2, 5–6, 29of wrongs, 176, 201–2

categorisation, 28–9, 42, 67, 369causation, 45–6, 110, 219–20, 403–4change of position, 207

Chitty, J, 58choses in action, 273, 280, 283, 286,

289–91, 304, 316circularity, 78, 315–16, 318, 373, 408civil law, 27, 31–2, 48, 77–9

and interpretation, 83–94civilian tradition, 32, 35, 48co-contractors, 308–9codification, 25, 27, 31–2, 39, 176coercion, 19, 116, 123, 125–7, 131–2,

135–7forcible, 126, 132–3

coercive remedies, 116, 123, 125–7, 132–3,137

cogency, 422collateral terms, 367, 372–3commercial contracts, 98, 103, 201, 352–3

negotiated, 246–7, 251, 253–4commitments, 141, 211–15, 217–18, 265,

370common fundamental mistake,

see common mistakecommon intention, 82, 91common law, 3, 29–31, 48–9, 288–91,

321–3, 325–7, 331–4and interpretation, 94–113mistake, 323, 369

common mistake, 319–25, 327, 329–39,365, 369, 373–375, 378

in The Great Peace, 319–39remedial flexibility, 334–8and third parties, 331–4

compensable loss, 148–9compensation, 14–15, 43, 47, 141, 158,

164–7, 172compensatory awards see compensatory

damagescompensatory damages, 43, 141, 143–5,

147, 149–50, 157–67competing policy interests, 329–30conceptual looseness, 89, 96, 99, 112–13conceptual tightness, 89–90, 94, 113conceptualisation, 28, 419conditions precedent, 110, 311–12, 321–2,

367implied, 321–3

conditions, 83–4, 154, 311–13, 316, 337–8,351–2, 370

contingent, 312conduct, 57–8, 203–4, 206–8, 212–17,

345–6, 349–53, 364–5buyers, 358inconsistent, 203–4, 206–8, 210, 224sellers, 358, 362unacceptable, 413

confidence, abuse of, 406, 410confidential relationships, 401, 409–10,

412, 415

432 Index



confidentiality, 149, 406, 409, 411conflict of interest, 383–4confusion, 48, 172, 231, 238, 255, 343–4,

362conscience, 54, 56, 72, 203, 290, 294, 388consecrating/disciplining distinction, 82,

86–91, 94–5, 98–9consecrating functions, 82–3, 86, 93–8,

100–2, 105, 107–9, 111–13consent, 19, 201–2, 251–2, 315–16,

369–71, 374, 391–2defective, 344, 369, 371, 376prior, 313, 315–16quality of, 252–3

consequential losses, 11–13, 15, 172–5,177–8, 180–2, 193–8

proof of, 176, 186, 197consideration,

compared to equitable assignments,272–78

and cultural criterion, 135–6doctrine of, 8, 51–75,116, 123, 126–8,

133–4, 166and morality of promising, 115–37

consolidation of law and equity, 99–100constitutive elements, 83–4construction, rules of, 98, 105–7constructive trusts, 272, 274, 278, 287,

294, 307, 392consumer contracts, 160, 192contingent conditions, 312contra proferentem rule, 89contract-creating events, 6–7

basic, 8–9, 21contract prices, 60–1, 176, 178, 191, 197contracting parties, 11, 18–19, 53, 69, 140,

150–3, 342contracts:

of assignment, 286–8, 291, 316commercial see commercial contractsconsumer, 160, 192employment, 250–1, 320, 328, 372of guarantee, 323–4of hiring, 227of sale, 37, 40–1, 47, 179–80, 184of service, 231–2

contractual documents, 349, 351–3, 355,364, 372–3, 375, 378

contractual duties, 11–12, 15, 306contractual obligations, 7–8, 16–17, 34–6,

92–3, 110, 304–5, 316–18see also obligationsnature of, 12–13primary, 151, 172

contractual prohibitions, 268, 285, 288contractual promises, 44, 54, 123, 127–8,

132, 209, 308–11see also promises

contractual relations, 225, 247, 280, 285direct, 285–7, 291, 307, 309, 312, 317

contractual rights, 229, 267, 285–9, 294–5,308–10, 314, 394

see also rightsbargained-for, 166

contributory negligence, 148, 245, 249–51conventions, 84–6, 89, 93, 164conveyances, 286–7, 309, 383Corbin, AL, 248, 254, 257, 302, 304–6corrective justice, 29, 47, 128–30cost of cure awards, 145–9counterclaims, 273, 275, 276counterproof, 420, 422court-ordered rights:

law of, 14–18orders to perform existing duty, 15–17orders to perform new duty, 17–18

court orders, 11, 15–18, 21–3covenants, 16, 55, 163, 187, 310

breach of, 313–15not to build, 192–4to repair, 187

criminal law, 3, 5, 7, 22–3culpa in contrahendo, 45cultural criterion, 118–19

and consideration, 135–7culture, 27–8, 118, 136

moral, 119, 135–6cure damages see cost of cure awards

damages, 9–18, 38–44, 140–54, 165–9,171–98, 256–7

compensatory, 43, 141, 143–4, 147,149–50, 155–67

expectation, 41, 43, 47, 56, 119, 139,222–4, 257

gain-based, 139, 193–4general, 173, 184general law of, 14, 44, 46and The Golden Victory, 195–7justification, 139–70law of, 10–14liquidated, 112, 189measure of, 12–13, 152mitigation of , 11, 13, 15, 148, 169,

181–2, 191, 193monetary, 122, 139, 143–4, 174nominal, 14, 17–18, 143, 162, 168,

194–5performance interest, 144–9, 155, 165punitive, 14, 17, 119quantification of, 45, 152, 182, 186,

191, 193reliance, 31, 44–6, 139, 222, 427remoteness rule of see remotenessand right to performance, 171–98special, 173, 184

Index 433



substantial, 149, 162, 165, 176, 180–1,187–90, 195–7

substitutive, 139, 165–9, 173, 175tort, 12vindicatory, 14

death, 57, 70, 258, 384, 428deceit, 176, 237–9, 265

contracting out of liability for, 239–54declarations of trust, 269, 271, 281

see also trustsdefamation, 2, 4, 34, 147defeasibility, 309, 314, 316–18defective consent, 344, 369, 371, 376defective goods, 41, 179–82defective performance, 147, 156, 181defences, 22–3, 141, 232–3, 238, 245,

249–50, 275–6justification, 232–4

deference, 396–7, 403deferential trust, 395–8, 403–5, 411, 414,

428breach of, 404, 412, 417

delivery, 177–8, 182–5, 300–1, 304late, 11, 174, 183time of, 179, 182

dependence, 57, 395–6, 400, 414–16, 425designations, 100–2, 106detached-formal objectivity, 355–6, 362detached objectivity, 348–9, 351–2detrimental action, 219–20, 224detrimental reliance, 73, 207, 210–11,

218–21, 224direct contractual relations, 285–7, 291,

309, 312, 317directors, 217, 299, 328, 390–1, 395disabilities, 364, 378, 394–5disadvantage, manifest, 407–8disciplining functions, 82–3, 88, 91, 93–5,

97–8, 100–2, 112–13disclaimers, 240, 247, 254, 256–9, 262–4discretion, 47, 274–5, 277, 322

equitable, 336–7remedial, 334–6

discrimination, 175dishonesty, 126, 241, 243, 246–7, 345

see also honestyDobbs, D, 249dominant parties, 399, 401, 418duress, 5, 9, 18–23, 37, 61–2, 371, 388,

413–14economic, 19, 60, 62, 231

duty:breach of, 11, 13, 163, 210, 316, 403ethical, 126, 133, 135fiduciary, 275, 328, 384, 390–1, 395,

403, 409–10of loyalty, 392, 403

moral, 63, 115, 117–18, 120, 122, 127,137

primary, 12–14, 18, 23, 147, 194special, 382, 392, 417

economic duress, 19, 60, 62, 231Edelman, J, 211effective anti-assignment clauses, 268, 288efficient breach, 16, 145Eisenberg, M, 396employees, 67–8, 99, 103, 175, 259, 360,

395employers, 99, 175, 189–91, 195, 226,

283–4, 395employment, 191, 320, 328, 372, 395, 398employment contracts, 250–1, 320, 328,

372enforcement, 55, 57–9, 65–6, 82–3, 247–8,

251–3, 273–4of exculpatory agreements/terms, 238,

242–4, 247–9, 251–3, 265of interpersonal morality, 115, 117of promises, 57, 211

enrichment, 71, 161, 216unjust, 2, 22–3, 63–4, 70–1, 161,

215–16, 335–6entire agreement clauses, 355, 373equal value, 56, 144, 155equitable:

assignees, 270, 274, 291, 293, 307assignments, 272–3, 283–318

apart from transfer, 286–9of benefits, 284–304compared to trusts, 272–8and contract terms, 296–304and discharge by performance,

296–304English approach, 270–2and insolvency, 273minimalist view, 304–17as transfers, 289–96

claims, 99, 206, 255discretion, 336–7estoppel, 71, 199, 201–3, 205–6,

208–12, 224Australia, 199, 221

fraud, 71, 380interests, 206, 271, 279, 292, 294,

305–9, 311jurisdiction, 95, 203, 323, 325, 331–2,

380, 401mistake, 369, 378rescission, 255–6, 266, 331set-offs, 275–6, 302wrongs, 175, 384

equity, 71–3, 270–3, 289–91, 293–5,306–9, 325–7, 334–6

consolidation of law and, 99–100

434 Index



courts of, 70, 95, 203, 206, 212, 306,360

rules of, 93, 289–90estoppel, 5, 7, 72–4

equitable, 71, 199, 201–3, 205–6,208–12, 224

promissory, 73, 200, 208, 211, 218, 260proprietary, 70–1, 73, 199–201, 204–5,

207, 210–18, 220–3and rights-creating events, 199–224substantive doctrines of, 199, 202, 218

ethical duties, 126, 133, 135events-based taxonomy, 202–3evidence, 97–8, 247–8, 251, 350–1,

358–60, 420–2, 424–6see also burden of proofof intention, 98, 355reliable, 55–6rules, 73, 90, 97–8, 350–1, 372

evidential burden/onus see burden of proofevidential presumptions, 423–4, 428exact performance, 146, 154, 270, 304exclusionary clauses, 105–8, 110exculpatory agreements, 237–53, 256

clear, 245, 254exculpatory categories, 386, 388, 429exculpatory terms, 238, 240–3, 246–7,

253–4, 256–61, 263–5clear, 244–5, 254–5, 258

expectancy, 43–5, 47, 383expectation damages, 41, 43, 47, 56, 119,

139, 223expectations, 47, 57–8, 208–10, 213–14,

222–3, 312–13reasonable, 68, 324, 326–7, 346, 350,

374expected value, 239, 253exploitation, 406–7, 414–15, 429–30express influence, 419extra-contractual misrepresentation, 240,

260extra-contractual representations, 240, 246,

254–5, 260, 266

factual advantages, 150–1, 155–6, 159–60,166–7, 169

factual harms, 156, 161–2, 167, 169factual losses, 166–8failure to perform, 12, 17, 116, 127, 140,

142, 144see also non-performance;

performancefair-dealing rules, 404, 417fairness, 12, 66, 106, 255, 326, 347, 404fallibility, 248, 253, 255fault, 19, 41, 93, 256, 322, 337, 369fiduciaries, 194, 381–3, 390–6, 401–3,

405–8, 411–13, 416–18

fiduciary:accountability, 380, 390–1, 395–6, 401,

415disloyalty, 392duties, 275, 328, 384, 390–1, 395, 403,

409–10breach of, 275, 390, 410

functions, 381–2, 384, 392, 416influence, 383, 386law, 328, 381, 384, 394, 401, 403,

409–10liability, 390, 392, 404obligations, 390–6, 398, 401–2, 409–12,

418, 430principle, 381, 387, 396, 400, 414rationale, 383, 388–9, 408, 419, 421,

427regulation, 386, 389–95, 405, 427

relational undue influence as,398–402

relations/relationships, 380–2, 384–5,390, 394, 401, 410–11, 417

responsibility, 393–4, 417, 425rules, 384, 405, 421, 427–8

relational undue influence as,389–405

status, 391, 394–5trust, 403, 405, 418

types, 394–8financial information, 246, 252, 263financial statements, 262–3Finn, P, 381Finnis, J, 351, 353Flannigan, R, 390–7, 402, 410, 425flexibility:

interpretive, 99–100remedial, 327, 334

force, 125–6, 133, 135, 228, 230, 271, 350physical, 125, 127threats of, 125

force majeure, 111–12, 334forfeitures, 105, 310, 313–15form contracts, 240–2, 247, 257, 352

see also boilerplateformal objectivity, 351–2formation of contracts, 37, 44–5, 83–4,

139–40and intention, 341–78and objectivity, 356–68see also offer and acceptance

fraud, 55, 71, 237–43, 245–53, 332–3,379–81

claims, 238–46, 248, 252, 258–60, 265unfounded, 248, 253, 265–6

clear and convincing evidence of, 244,248, 251, 253

equitable, 71, 380freedom, 19, 38, 44, 48, 110, 125, 202

Index 435



French law, 32, 36, 81, 101, 103, 108–13and interpretation, 83–94

Fried, C, 116, 121frustration, 9, 18, 95, 109–11, 113,

329–30, 333–5Fuller, L, 42–4, 47functionalism, 82fundamental mistakes,

see common mistake

gain-based damages, 139, 193–4Gaius, 27–9, 33, 35general damages, 14, 44, 46, 173, 184general rules, 2, 9, 21, 39–40, 43–6, 250–2generality, 16, 38, 42, 52German law, 3, 30–5, 37–40, 42, 44, 46–7gift promises, 58, 120gifts, 22, 37, 53, 120, 238, 272, 398–400

executed, 7, 22Gilbert, J, 54, 56, 63–4, 66Gilbert, M, 123–5Gilmore, G, 43good faith, 61, 77–8, 92–3, 264Goode, RM, 310–11goods, 15, 176–80, 182–6, 332–3, 362–3,

366, 409carriage of, 184–6defective, 179–82

gross negligence, 264–5guarantees, 92, 258, 323, 354

contracts of, 323–4

hardening of categories, 31, 38, 42, 48–9hardship, 97, 385harms, 6, 164, 207–8, 223–4, 328, 348,

384factual, 156, 161–2, 167, 169physical, 158–9, 161, 165, 237

Hart, HLA, 124–5, 369–71history, 25, 32–3, 42, 44, 56, 74, 147Holdsworth, W, 228, 235, 270Holmes, OW, 16, 141–2, 156honesty, 241, 243, 252–3, 328, 343, 373,

401see also dishonesty

horizontal borders, 2–7, 34

identity,mistake of, 104, 332, 344, 366–8, 375–8

implied-in-fact terms, 18, 108–9implied-in-law terms, 9, 12, 18, 108–9implied intentions, 333implied terms, 77–8, 95, 108–9, 152–3,

321, 333–4, 338–9impugned benefits, 383, 405, 418impugned transactions, 381, 386–7, 402–4,

406–7, 423, 426, 430

inaccuracy, 254, 256, 259–60, 266inadvertent misrepresentation, 237, 266

contracting out of liability for, 254–60incapacity, 18–19, 22–3, 385

see also capacityincentives, 256, 332, 367, 383–4inconsistent conduct, 203–8, 210, 224incontestability clauses, 239incorporation of terms, 8–9, 15, 85, 365indemnities, 152, 178, 245, 293induced mistakes, 371, 375inducement, 204–7, 210, 218–19, 238–9,

241, 252–3, 259–60of a change of position, 207–8

inducing breach of contract, 226–7, 230–1inferences, 24, 42, 385, 419–20, 426

permissible, 382, 388–9, 419–21, 424,427

influence:express, 419fiduciary, 383, 386limited-access relations of, 403, 407,

414, 416, 428presumptions of, 424–5relationships of, 381, 400, 411, 417, 425special, 381, 386, 403–4, 426special relations of, 381, 386, 419, 423undue see undue influence

informal assignments, 286–7information, 29, 149, 208, 238, 257, 260–6

financial, 246, 252, 263supplying of, 238, 259, 261–4

infringements of rights, 149–50, 166,172–3, 193, 195, 202–3

see also breach; rightsinjury, 11, 43–5, 111, 129–30, 165, 227,

232injustice, 32, 70, 97, 154, 187, 254, 276innocent misrepresentation, 255–7, 375innocent third parties, 331–3, 335, 366insolvency and equitable assignments, 273insurance, 239, 262insurance contracts, 239insurers, 67, 70, 239, 275, 277–8intention, 82–3, 85–99, 101–2, 341–4,

346–7, 351–3, 355–6actual, 45, 90, 97, 99, 109, 344–5common, 82, 91evidence of, 98, 355and formation of contracts, 341–78implied, 333objective, 89, 95–102, 108, 112–13, 345,

361real, 55, 58, 95, 97, 103, 345–6, 355reasonable, 96, 109subjective, 89–90, 94–6, 98, 113, 345–6,

361–4subjective conception of, 86–7

436 Index



and vitiation, 341–78interests, 6, 172, 205–7, 291, 293–4,

381–5, 390–3best, 429–30equitable, 206, 271, 279, 292, 294,

305–9, 311personal, 239, 241–2, 252, 384, 393

interference with economic relations, 229,231

intermediaries, 395–7interpersonal morality, 115, 117–18, 120–1

enforcement of, 115, 117interpretation, 8–9, 77–114, 350–1

civil law, 83–94classic canons of, 89, 94common law, 94–113contextual, 91, 353French law, 83–94general principles of, 105–6literal, 91, 97, 99, 281, 299, 353, 360objective, 98, 342, 372, 377

interpretive flexibility, 99–100invited reliance, 238, 261–4, 266

judicial activism, 225, 234justice:

corrective, 29, 47, 128–30general considerations of, 63, 66–7

justifiable reliance, 249, 255justification defence, 232–4

known mistakes as to terms, 343, 358–9,378

Langdell, CC, 26late delivery, 11, 174, 183–4latent ambiguity, 363legal categories, 2, 5–6, 29

see also categorieslegitimate expectations see reasonable

expectationsliability, 224–6, 228–34, 237–9, 241–7,

251–61, 263–5accessory, 225, 229contracting out of, 237–66fiduciary, 390, 392, 404negligence, 238, 261–4strict, 256, 259, 384, 392

‘limited-access’ abstraction, 393–4see also access to assets

limited-access arrangements, 386, 393, 398,401–2, 428–9

limited-access relations of influence, 403,407, 414, 416, 428

limits of contract, 1–24classificatory criteria, 6–8

liquidated damages, 112, 189

literal interpretation, 91, 97, 99, 281, 353,360

literalism, 91, 96–9Locke, J, 126, 129–30Lord Cairns’ Act, 187losses, 43–7, 142–5, 147–9, 158–69, 172–9,

181–7, 191–3actual, 47, 171, 177–8, 182, 188, 195–6compensable, 148–9consequential, 11–13, 15, 172–5, 177–8,

180–2, 193–8factual, 166–8non-pecuniary, 161, 164pecuniary, 45, 161proof of, 173–4recovery of see recoveryreliance, 217, 222–3

loyalty, 353, 382, 392, 403

MacMillan, C, 327–8malice, 228, 230manifest disadvantage, 407–8mapping contract law, 25–6, 28, 30–1,

48–9, 52–3market price, 176–7, 182–4, 197, 360market value, 160–1, 176–81, 183, 185,

190, 196, 198markets, 33, 164, 177, 179, 183, 187,

191–2marriage contracts, 53, 85Marshall, OR, 286–7master and servant, 226–27, 232McFarlane, B, 200, 204–5, 307McMeel, G, 107meaning, natural, 99–100, 102–3, 105–7merger provisions, 240, 248, 254, 256–8minors, 20, 39, 262misrepresentation, 18–23, 237–9, 251–2,

344, 357–9, 375–8see also representationsextra-contractual, 240, 260inadvertent, 237, 254–60, 266innocent, 255–7, 375intentional, 19, 247negligent, 237–8, 251, 254, 357reckless, 246–7

misstatement, 237–8, 258, 260, 264negligent, 237, 242, 255, 257–65

mistake, 319–21, 323–5, 330–1, 341–78categories of, 344, 369common law, 323, 369common, 319–25, 327, 329–39, 365,

369, 373–375, 378contracts void for, 373equitable, 369, 378of fact, 344, 371–2, 377of identity see mistaken identityinduced, 371, 375

Index 437



mutual, 257, 320, 363, 369and objectivity, 372–3operative common, 321, 332, 334shared, 337, 375subjective, 343, 363of terms, 342–4, 355, 357, 360, 371–3,

376–7known, 343, 358, 378operative, 356, 374

unilateral, 323, 332, 347, 360–1, 369,374–5

mistaken assumptions, 368–76mistaken identity, 104, 332, 344, 366–8,

375–8mitigation, 11, 13, 15, 148, 169, 181–2,

191, 193modification of contracts, 41, 59, 61–2monetary damages, 122, 139, 143–4, 174money, 16, 62–3, 142–6, 160, 163–5,

174–5, 216–17moral:

agents, 115–21, 127–8, 130–2, 134–5,137

behaviour, 119, 135–6concepts, 38, 123–4culture, 119, 135–6duties, 63, 115, 117–18, 120, 122, 127,

137obligations, 6, 120, 133, 136permissibility, 121–2, 127–8principles, 57, 119, 121, 132relationships, 118, 124, 129remedies, 122–3, 133, 135, 137rights, 116, 123–8, 131–3, 136–7

strong, 127, 132–4weak, 127–8

morality, 20–1, 115–17, 119–27, 131–3,135–7, 233–4

interpersonal see interpersonal moralitypromissory, 115–37

mutual mistake, 257, 320, 363, 369

natural meaning, 99–100, 102–3, 105–7negligence, 67–8, 104, 107, 208, 238, 242,

261, 264contributory, 148, 245, 249–51gross, 264–5liability, 238, 261–4tort of, 261–2

negligent misrepresentation, 237–8, 251,254, 357

negligent misstatement, 237, 242, 255,257–65

negotiated commercial contracts, 246–7,251, 253–4

negotiations, pre-contractual, 361, 371nemo dat, 295–6, 331

nineteenth century legal thought, 26, 32–3,38, 64, 93, 225

no-conflict rules, 384, 392, 404, 418no-profit rules, 392, 403nominal damages, 14, 17–18, 143, 162,

168, 194–5non-assignability, 304, 315

of personal obligations, 304–6non-assignable rights, 279–80non-assignment provisions, 279, 305

see also anti-assignment clausesnon-contractual obligations, 9, 36, 39non-contractual reliance, 71non-delivery, 176–9, 182–4, 198, 375non est factum, 364–5non-pecuniary losses, 161, 164non-performance:

see also failure to perform; performancerights arising from, 10–18

non-profit-making assets, 174non-reliance clauses, 240–1, 244–5, 254novation, 289, 291, 307, 318nuisance, 2, 4, 12, 16

objective:agreements, 356, 359–60, 362, 364intention, 89, 95–102, 108, 112–13, 345,

361conceptual looseness of, 96, 99, 113

interpretation, 98, 342, 372, 377law, 88, 112test, 342–4, 347–8, 363, 376

objectivity, 98, 342–51actor, 349–51, 364–5and contract formation, 356–68defending, 344–55defining, 348–55detached, 348–9, 351–2detached-formal, 355–6, 362formal, 351–2justification for, 345–7and mistake, 372–3observer-contextual, 355–6, 362, 367,

377–8and voluntariness, 347–8

obligation-creating events, 6–7obligations, 2–8, 33–41, 84–90, 139–42,

149–53, 199–204, 267–82cartography of, 25–49and factual structure, 402–5fiduciary, 390–6, 398, 401–2, 409–12,

418, 430moral, 6, 120, 133non-contractual, 9, 36, 39personal, 299, 304–5, 315primary, 3–6, 10, 34, 139, 156, 159promissory, 123, 133tortious, 38–9, 411

438 Index



voluntary, 218, 346–8, 350obligees, 268, 271, 297–301, 304–5, 310,

312–13, 391–2obligors, 218, 280–1, 284–5, 287–9,

291–309, 311–13, 315–18observer-contextual objectivity, 355–6, 362,

367, 377–8offer and acceptance, 8–9, 18, 37, 40, 84,

210, 300, 359, 363, 370–3onerous terms, 92, 95, 105–7, 113, 353opportunism, 391, 394, 396, 401–2, 405,

428–9overprotection of contracts, 225–35ownership, 37, 116, 126–7, 201, 214, 294,

368–9

Paley, W, 57–8Pandektensystem, 35–6parol evidence rule, 97–100, 105, 254,

256–9, 266pecuniary losses, 45, 161Peden, E, 99Peel, E, 107, 342–3Perdue, W, 42–4, 47performance, 15–18, 139–60, 166–75,

185–7, 195–8, 299–306agreed exchange of, 330defective, 147, 156, 181discharge by, 296–304exact, 146, 154, 270, 304factual changes to, 297interest, 141, 155, 157, 165, 172

damages, 144–9, 155, 165in lieu of, 40–1, 186limits, 149–57right to and damages, 171–98specific, 9, 15–18, 24, 41, 56, 67, 119,

142–4, 158, 172, 186–7, 222,280, 290–3, 311–13, 318, 336,360–2

standing to demand, 125, 127, 132–3,137, 149

substitute, 11–13, 16value of, 12, 187, 192

perjury, 55permissibility, moral, 121–2, 127–8permissible inferences, 382, 388–9, 419–21,

424, 427personal obligations, 299, 315

non-assignability of, 304–6persuasion, 233, 385, 387, 413, 420, 426,

429unfair, 387, 429

persuasive burden, 385, 420, 422physical harms, 158–9, 161, 165, 237Pineau, J, 94policy interests, competing, 329–30policy presumptions, 384, 421–2, 424, 426

policy reasons, 68, 285, 384see also public policy

possibility regulation, 392–3, 423post-notice equities, 276–8Pound, R, 29, 35–6pre-contractual negotiations, 361, 371precedents, 51, 81, 97, 100, 225, 325–7

condition, 110, 311–12, 321–2, 367pressure, 19, 61–2, 231, 319, 386presumed undue influence, 382, 387,

417–18, 421presumptions, 109, 219–20, 368, 382–90,

400–1, 406–12, 416–29evidential, 423–4, 428policy, 384, 421–2, 424, 426probability, 421–2, 424of undue influence, 382–6, 390, 398,

406–12, 417, 419–25, 429price, 59–60, 140, 178–9, 182, 298, 356–8,

361primary contractual obligations, 151, 172primary duties, 12–14, 18, 23, 147, 194primary obligations, 3–6, 10, 34, 139, 156,

159primary rights, 122, 147, 172, 175, 202–3,

207prior consent, 313, 315–16private law, 15–16, 23, 25–8, 36–8, 48–9,

128–9, 139–40private ordering, 246, 248, 253, 265private rights of action, 116, 128–33,

136–7, 306private wrongs, 53, 64privity, 63, 65, 67–70, 74, 190, 226, 231–2probability regulation, 392–3, 423profit-making assets, 174profits, 19, 45–6, 103, 178, 181, 183,

192–4promises, 20–3, 55–9, 62–4, 115–23,

125–8, 133–7breach of, 47, 121, 127, 132, 207enforcement of, 57, 211and estoppel 211–18gift, 58, 120morality of see promissory moralityordinary, 127, 132–3rash, 54, 56and rights-creating events, 209–24

promisors, 56–8, 115–16, 120, 125–6,131–3, 135–7, 349

promissory:estoppel, 73, 200, 208, 211, 218, 260morality, 115–37obligations, 123, 133remedies, 222–4rights, 120–3

proof:see also evidence

Index 439



burden of see burden of proofof consequential loss, 176, 186, 197

property, 53–4, 147, 199–201, 204–7, 217,362

and obligations, 267–82proprietary estoppel, 70–1, 73, 199–201,

204–5, 207, 210–18, 220–3public order, 84–5, 93public policy, 84, 246, 252, 288, 381–2,

385–6, 405–8puffery, 249–51punitive damages, 14, 17, 119

quality:of consent, 252–3of life, 159, 161, 164–5

quantification of damages see damages,quantification of

Raz, J, 348, 353reasonable expectations, 68, 324, 326–7,

346, 350, 374reasonable intentions, 96, 109reasonable person, 96, 208–9, 213, 249,

345, 349, 351reasonableness, 191, 212, 343, 350, 373reckless misrepresentation, 246–7rectification, 91, 97–9, 309, 355, 372–3,

375–6redress, 14, 129–33, 239, 241–2, 253,

255–6, 266regulation:

fiduciary, 386, 389–91, 394–5, 398, 402,405, 427

possibility, 392–3, 423probability, 392–3, 423

relational undue influence, 385–90, 429–30and burden of proof, 419–27claims, 402, 412, 416, 418, 420as fiduciary regulation, 398–402as fiduciary rule, 389–405law of, 390, 395, 412, 417National Westminster Bank plc v

Morgan, 405–12nature of, 401, 410Royal Bank of Scotland plc v Etridge

(No 2), 412–19reliance, 42–7, 70–3, 207–13, 215–24,

244–7, 249–52, 261–6beneficial, 208damages, 31, 44–6, 139, 427detrimental, 73, 207, 210–11, 218–21,

224interest, 216, 222

and damages, 42–8invited, 238, 261–4, 266justifiable, 249, 255

losses, 217, 222–3non-contractual, 71unexpected, 251unjustifiable, 249–50unreasonable, 249, 251

remedial discretion, 334–6remedial flexibility, 327, 334remedies:

coercive, 116, 123, 125–7, 132–3, 137moral, 122–3, 133, 135, 137promissory conception, 222–4

remoteness, 11, 13, 15, 23, 111, 180–1renegotiation, 59–61replacement, 146, 158, 174, 181, 198representations, 212, 215–16, 219, 221–2,

239–47, 249–51, 254–60see also misrepresentationextra-contractual, 240, 246, 254–5, 260,

266repudiation, 41, 186, 195–6res extincta, 327, 329, 331–2res ipsa loquitur, 384, 419–23, 425res sua, 327, 329, 331–2rescission, 255–7, 322, 324, 334–5, 337–8

equitable, 255–6, 266, 331responsibility, 13, 18, 21–3, 39, 237–8,

382–4, 401–3fiduciary, 393–4, 417, 425

Restatement (Second) of Contract, 211,218, 254, 330

restitution, 4, 12, 71, 215, 255–7, 266see also unjust enrichment

restrictive covenants see covenantsrighteousness, 383, 392, 406, 418rights:

of action see private rights of actionnon-assignable, 279–80to performance

and damages, 171–98identification, 140–1

primary, 122, 147, 172, 175, 202–3,207

secondary, 122, 202of set-off, 273subjective, 40subrogation, 70, 277third party, 70, 327, 333, 368

rights-creating events:and estoppel, 199–224and promises, 209–24and wrongs, 201–9

Ripstein, A, 156risk allocation, 323, 339rules:

of construction, 98, 105–7of equity, 93, 289–90fiduciary, 384, 389, 405, 421, 427–8no-conflict, 384, 392, 418

440 Index



no-profit, 392, 403signature, 351–2, 365, 372

sale, contracts of, 37, 40–1, 47, 179–80,184

sale of goods, 109, 176–84, 409sale of land, 186–7Samuel, G, 28sanctity of contract, 229, 324, 326–7, 330Savigny, FC von, 33, 36–8Sayre, FB, 229Schlag, P, 25secondary rights, 122, 202servants, 53, 58, 66, 226–7, 232, 328, 395service, contracts of, 231–2set-off, 271, 273, 276–8, 302–4, 318

equitable, 275–6, 302rights of, 273

shared mistake, 337, 375Shiffrin, S, 115–22, 127–8, 130–7signatures, 216, 350, 352–3, 364, 366

rule, 351–2, 365, 372special capacity, 382, 398–9special damages, 173, 184special duties, 382, 392, 417special influence, 381, 386, 403–4, 419,

423,426specific relief, 16, 143, 146, 148, 155, 159,

192specific performance, 9, 15–18, 24, 41, 56,

67, 119, 142–4, 158, 172, 186–7,222, 280, 290–3, 311–13, 318,336, 360–2

standard form contracts see form contractsstanding, 53, 125, 127–8, 132–3status-based relations, 381, 424Statute of Frauds, 55, 87statutory assignment, 270, 278, 283, 289,

291, 293, 301strangers, 63, 286, 288, 313–14, 316, 412strict liability, 256, 259, 384, 392strict rights, 72–3strong evidence, 242, 248, 250sub-buyers, 177, 179sub-sale contracts, 177–80, 183–4subcontractors, 60–1subjective intention, 89–90, 94–6, 98, 113,

345–6, 361–4subjective mistake, 343, 363subjective right, 40subjective trust, 394subjectivity, 346, 348, 355–6, 364subordinate parties, 382, 399, 403subrogation, 70, 275, 277–8subservient parties, 418, 425substantial damages, 149, 162, 165, 176,

180–1, 187–90, 195–7substitute performance, 11–13, 16

substitutes, 10–12, 82, 122, 132, 157–8,167

substitution, 158, 165, 172, 174, 192, 270substitutive damages, 139, 165–9, 173, 175surrogate assignment, 277suspicious transactions, 422, 425–6swindlers see fraud

taxonomy, 6, 26–9, 31–4, 36–8, 48–9, 199,368–71

events-based, 202–3tenants, 74, 187, 209, 213–14, 216, 354,

359tendering contracts, 77, 109terms, 105–10, 152–3, 251–4, 295–8,

334–8, 356–61, 371–7clear exculpatory, 244–5, 254–5, 258collateral, 367, 372–3and equitable assignments, 296–304exculpatory, 238, 240–3, 246–7, 253–4,

256–61, 263–5implied-in-fact, 18, 108–9implied-in-law, 9, 12, 18, 108–9implied, 77–8, 95, 108–9, 152–3, 321,

333–4, 338–9imposed, 322, 334–7incorporation of, 8–9, 15, 85, 365invariability of, 296–304mistakes as to, 342–4, 355, 357,

359,360, 371–3, 376–7onerous, 92, 95, 105–7, 113, 353variability of, 296–304

Terré, F, 84tertium comparationis, 79, 82Thel, S, 211, 218third parties, 69, 145–7, 186–90, 280–3,

301–3assignees, 284–5, 317and common mistake, 331–4innocent, 331–3, 335, 366rights, 70, 327, 333, 368

threats, 21, 61–2, 203, 312–13, 370of force, 125unlawful, 413

Tolhurst, G, 267, 271, 280, 286–7,290–302, 305–7, 315–17

tort, 2, 5, 13–14, 128, 131, 147, 201–2,224–5, 228–9, 237–8, 255–6

actions/claims, 13, 252, 255, 258–60damages, 12–13inducing breach of contract, 225–34interference with economic relations,

229, 231Lumley v Gye, 225–35

tortious obligations, 38–9, 411transactions:

arm’s-length, 388–9

Index 441



impugned, 381, 386–7, 402–4, 406–7,423, 426, 430

suspicious, 422, 425–6transfer, 7–9, 20–1, 37–8, 269–72, 278–80,

305–7and equitable assignments, 286–96

transferability, 317–18transparency, 93, 107, 118, 133, 135Treitel, G, 56, 342, 425trespass, 2–4, 7, 10, 12, 16, 34, 64trust:

breach of, 274–5, 404declarations of, 269, 271, 281deferential see deferential trustfiduciary, 394–5, 398, 403, 405, 418property, 277, 404subjective, 394relationship of, 397, 401, 403, 415–16vigilant, 395, 397–8, 403–5

trust and confidence, 418–19abuse of, 388, 414–15

trustees, 188, 269, 271, 274–7, 279–80,287, 305

trusts, 22, 267–72and assignments, 278–82bare, 271–2, 274–5, 277–8beneficiaries, 269, 273–6, 279, 282constructive, 272, 274, 278, 287, 294,

307, 392and equitable assignments, 272–8

unconscionability, 24, 37, 71, 106–7,203–4, 414, 430

unconscionable dealing, 384, 388–9, 414,429–30

undertakings, 39, 83, 140–2, 151–3, 156–7,260–3

undue influence, 18–23, 399–450actual, 382, 415, 418changes in law, 386–9claims, 386, 402, 408, 411–12, 415–16,

418, 420–1presumed, 382, 387, 417–18, 421presumptions of, 382–6, 390, 398,

406–12, 417, 419–25, 429relational see relational undue influencetypes, 380–6

unfair persuasion, 387, 429unfairness, 80, 106–7, 388, 396unilateral mistake, 323, 332, 347, 360–1,

369, 374–5unjust enrichment, 2, 35, 39, 63–4, 70–1,

161, 215–16, 369

law of, 4, 23, 335–6unjustifiable reliance, 249–50unjustified enrichment see unjust

enrichmentunlawful threats, 413unreasonable reliance, 249, 251unreasonableness, 251, 361

value:of assets, 160–1, 381, 402equal, 56, 144, 155expected, 239, 253of performance, 12, 187, 192

variations, 59–60, 239, 296–7, 302, 309,318

vertical borders of contract, 2–7, 9, 21, 34,42–4, 48–9

vesting, 250, 284, 292, 309, 313, 315victimisation, 386–7, 405–8, 429–30vigilant trust, 395, 397–8, 403–5vindicatory damages, 14vitiation, 39

and intention, 341–78Vogenauer, S, 79, 87, 90void ab initio, 84, 321–2, 334void contracts, 342, 344, 376voluntariness, 210, 218

and objectivity, 347–8voluntary obligations, 218, 346–8, 350vulnerability, 380, 394, 396, 414, 416, 430

waivers, 70, 219, 239, 242, 252–3, 314–15warranties, 242, 254, 256, 331

breach of, 242, 254Weinrib, E, 11, 154–7, 347Williston, S, 43, 248, 254, 257wrongful act, 228, 381wrongs, 53, 128–32, 139, 197

categories of, 176, 201–2and damages, 172–6equitable, 175, 384inconsistent conduct, 203–7inducing a change of position, 207–9law of, 202–3, 207, 210, 224private, 53, 64and rights-creating events, 201–9

Yorio, E, 211, 218

Zimmermann, R, 3Zipursky, B, 128–31

442 Index


Documents

[Jason_W._Neyers,_Richard_Bronaugh,_Stephen_G._A._(BookFi.org).pdf